National Collaborating Centre for Women’s and Children’s Health

Diabetes in pregnancy
management of diabetes and its complications from preconception to the postnatal period

Clinical Guideline

March 2008 (revised reprint July 2008)

Funded to produce guidelines for the NHS by NICE

Diabetes in pregnancy
management of diabetes and its complications from preconception to the postnatal period
National Collaborating Centre for Women’s and Children’s Health Commissioned by the National Institute for Health and Clinical Excellence

March 2008 (revised reprint July 2008)

RCOG Press

Published by the RCOG Press at the Royal College of Obstetricians and Gynaecologists, 27 Sussex Place, Regent’s Park, London NW1 4RG www.rcog.org.uk Registered charity no. 213280 First published March 2008, revised reprint July 2008 © 2008 National Collaborating Centre for Women’s and Children’s Health

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page. The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use. While every effort has been made to ensure the accuracy of the information contained within this publication, the publisher can give no guarantee for information about drug dosage and application thereof contained in this book. In every individual case the respective user must check current indications and accuracy by consulting other pharmaceutical literature and following the guidelines laid down by the manufacturers of specific products and the relevant authorities in the country in which they are practising. ISBN 978-1-904752-47-9 NCC-WCH Editor: Andrew Welsh Original design: FiSH Books, London Typesetting: Andrew Welsh Proofreading: Elisabeth Rees Evans Index: Jan Ross (Merrall-Ross (Wales) Ltd) Printed by Henry Ling Ltd, The Dorset Press, Dorchester DT1 1HD

Contents
Guideline Development Group membership and acknowledgements Guideline Development Group External advisers Guideline Review Panel Acknowledgements Stakeholder organisations Abbreviations Glossary of terms 1 Introduction 1.1 Diabetes in pregnancy 1.2 Aim of the guideline 1.3 Areas outside the remit of the guideline 1.4 For whom is the guideline intended? 1.5 Who has developed the guideline? 1.6 Other relevant documents 1.7 Guideline methodology 1.8 Stakeholder involvement in the guideline development process 1.9 Schedule for updating the guideline 2 Summary of recommendations and algorithm 2.1 Key priorities for implementation (key recommendations) 2.2 Summary of recommendations 2.3 Key priorities for research 2.4 Summary of research recommendations 2.5 Algorithm 3 Preconception care 3.1 Outcomes and risks for the woman and baby 3.2 The importance of planning pregnancy and the role of contraception 3.3 Diet, dietary supplements, body weight and exercise 3.4 Target ranges for blood glucose in the preconception period 3.5 Monitoring blood glucose and ketones in the preconception period 3.6 The safety of medications for diabetes before and during pregnancy 3.7 The safety of medications for diabetic complications before and during pregnancy 3.8 Removing barriers to the uptake of preconception care and when to offer information 3.9 Cost-effectiveness of self-management programmes 3.10 Retinal assessment in the preconception period 3.11 Renal assessment in the preconception period 4 Gestational diabetes 4.1 Risk factors for gestational diabetes 4.2 Diagnosis of gestational diabetes 4.3 Screening and treatment for gestational diabetes 5 Antenatal care 5.1 Target ranges for blood glucose during pregnancy 5.2 Monitoring blood glucose and ketones during pregnancy 5.3 Management of diabetes during pregnancy 5.4 Retinal assessment during pregnancy 5.5 Renal assessment during pregnancy 5.6 Screening for congenital malformations 5.7 Monitoring fetal growth and wellbeing 5.8 Timetable of antenatal appointments 5.9 Preterm labour in women with diabetes v v vi vi vi vi x xii 1 1 1 2 2 2 3 4 8 8 9 9 10 18 19 23 28 28 30 33 36 41 43 49 53 57 59 59 60 60 64 66 78 78 82 84 88 94 96 102 107 113 iii

Diabetes in pregnancy 6 Intrapartum care 115 6.2 Prevention and assessment of neonatal hypoglycaemia 136 8 Postnatal care 143 8.1 Breastfeeding and effects on glycaemic control 143 8.1 Initial assessment and criteria for admission to intensive or special care 130 7.1 Timing and mode of birth 115 6.3 Glycaemic control during labour and birth 125 7 Neonatal care 130 7.2 Information and follow-up after birth 146 Appendix A Declarations of interest 153 Appendix B Clinical questions 156 Appendix C Cost-effectiveness of self-management programmes for women with diabetes who are planning a pregnancy 158 Appendix D Cost-effectiveness of screening. diagnosis and treatment for gestational diabetes 165 Appendix E Cost-effectiveness of screening for congenital cardiac malformations 191 References 200 Index 213 Search strategies CD-ROM Evidence tables CD-ROM iv .2 Analgesia and anaesthesia 122 6.

Building Schools for the Future. London School of Hygiene & Tropical Medicine (formerly Research Fellow.Guideline Development Group membership and acknowledgements Guideline Development Group GDG member Dominique Acolet (until October 2007) Lynne Carney Anne Dornhorst Robert Fraser Roger Gadsby Jane Hawdon (from October 2007) Richard Holt Ann Parker Nickey Tomkins Stephen Walkinshaw Jackie Webb Saiyyidah Zaidi Job title and affiliation Clinical Director (Perinatal Epidemiology). University of Sheffield General Practitioner. NCC-WCH Research Fellow. London Job title and affiliation Senior Work Programme Coordinator. NCC-WCH Senior Health Economist. and Senior Lecturer in Primary Care. NCC-WCH Health Economist. NCC-WCH Research Fellow. University of Southampton Diabetes Advisory Midwife. University of Warwick Consultant Neonatologist. London Development Officer. Kent Consultant in Maternal and Fetal Medicine. Medway Primary Care Trust. Fair Share Trust. Royal Shrewsbury Hospital Advanced Nurse Practitioner in Diabetes/Diabetes Specialist Midwife. Hammersmith Hospital. Nuneaton. NCC-WCH Area of expertise Neonatal paediatrician Patient/carer representative Diabetes physician Obstetrician (GDG chair) General practitioner Neonatal paediatrician Diabetes physician Diabetes specialist midwife Diabetes nurse practitioner Obstetrician Diabetes specialist nurse Patient/carer representative NCC-WCH staff Paula Broughton-Palmer Michael Corkett Anthony Danso-Appiah Paul Jacklin Lorelei Jones Moira Mugglestone Jeffrey Round Anuradha Sekhri Area of expertise Guideline methodologist Guideline methodologist Guideline methodologist Guideline methodologist Guideline methodologist Guideline methodologist (NCC-WCH project director) Guideline methodologist Guideline methodologist v . Liverpool Women’s Hospital Diabetes Specialist Nurse Manager. University College Hospitals London NHS Foundation Trust Reader of Endocrinology and Metabolism. Confidential Enquiry into Maternal and Child Health (CEMACH). Birmingham Programme Manager. Community Foundation for Shropshire and Telford Consultant Physician and Honorary Senior Lecturer in Metabolic Medicine. NCC-WCH) Deputy Director. NCC-WCH Senior Information Specialist. London Reader in Obstetrics and Gynaecology. NCC-WCH Research Fellow. Heart of England NHS Foundation Trust.

and Honorary Consultant Physician. Newcastle Acute Hospitals NHS Trust and Newcastle Primary Care Trust Area of expertise Obstetric analgesia and anaesthesia CEMACH Diabetes in Pregnancy programme Ophthalmologist with an interest in diabetic retinopathy Diabetes physician with an interest in diabetic retinopathy Guideline Review Panel GRP member Mike Drummond Graham Archard Catherine Arkley Karen Cowley David Gillen Barry Stables Job title and affiliation Director. Diabetes in pregnancy: are we providing the best care? Findings of a national enquiry: England. London: CEMACH. Anita Fitzgerald.Diabetes in pregnancy External advisers External adviser Anita Holdcroft Jo Modder Peter Scanlon Roy Taylor Job title and affiliation Reader in Anaesthesia. Alyson Huntley. Sjokvist Garcia-Stewart. Jiri Chard. University of York (GRP Chair) General Practitioner. University of Newcastle. Imperial College London. London and Consultant Obstetrician. Stakeholder organisations Action on Pre-eclampsia Addenbrooke’s NHS Trust Adur Arun and Worthing PCT Afiya Trust Airedale General Hospital All Wales Dietetic Advisory Committee Anglesey Local Health Board Ashfield and Mansfield District PCT Association for Improvements in Maternity Services (AIMS) Association of British Clinical Diabetologists Association of British Health-Care Industries Association of Clinical Biochemists Association of Radical Midwives Association of the British Pharmaceuticals Industry (ABPI) AstraZeneca UK Baby Lifeline Barnsley Acute Trust Barnsley PCT Bedfordshire PCT vi . we thank CEMACH for allowing us to reproduce in our glossary terms from the glossary published in Confidential Enquiry into Maternal and Child Health. Finally. Dorset Lay Member Practice Development Nurse. and Honorary Consultant Anaesthetist at Chelsea and Westminster Hospital. CEMACH. Diabetic Retinopathy Screening Programme (England). York Medical Director. University College London Hospitals NHS Trust National Co-ordinator. Debbie Pledge and Jane Tuckerman at the NCC-WCH. 2007. Wyeth Pharmaceuticals Lay Member Acknowledgements Additional support was received from: Anna Burt. London Clinical Director (Obstetrics). Centre for Health Economics. Wales and Northern Ireland. Consultant Ophthalmologist. Gloucestershire Eye Unit and Oxford Eye Hospital Professor of Medicine and Metabolism. We also thank the Patient and Public Involvement Programme (PPIP) of the National Institute for Health and Clinical Excellence (NICE) whose glossary was adapted for use in this guideline.

Guideline Development Group membership and acknowledgements Birmingham Women’s Healthcare Trust Blaenau Gwent Local Health Board BLISS – the premature baby charity Bournemouth and Poole PCT Bradford & Airdale PCT Bradford Hospitals NHS Trust Brighton & Sussex University Hospitals Trust Bristol Health Services Plan British Association for Counselling and Psychotherapy British Association of Perinatal Medicine British Dietetic Association British Maternal and Fetal Medicine Society British National Formulary (BNF) British Psychological Society Buckinghamshire Hospitals NHS Trust Calderdale PCT CASPE CEMACH City and Hackney PCT City Hospitals Sunderland NHS Trust Cochrane Pregnancy & Childbirth Group Commission for Social Care Inspection Connecting for Health Conwy & Denbighshire NHS Trust Co-operative Pharmacy Association County Durham and Darlington Acute Hospital Trust Coventry and Warwickshire Cardiac Network Craven Harrogate & Rural District PCT Croydon PCT Cytyc UK Daiichi Sankyo UK Department of Health Department of Health. Social Security and Public Safety of Northern Ireland Derbyshire Mental Health Services NHS Trust Det Norske Veritas – NHSLA Schemes Diabetes UK Doncaster PCT East and North Herts PCT East Sussex Hospitals NHS Trust Echo UK (the tiny tickers charity) Eli Lilly and Company English National Forum of LSA Midwifery Officers Evidence Based Midwifery Network Faculty of Public Health Fibroid Network Charity Gloucestershire Hospitals NHS Trust Guy’s and St Thomas’ NHS Foundation Trust Healthcare Commission Heart of England NHS Foundation Trust HemoCue Ltd Independent Midwives Association Institute of Biomedical Science Institute of Health and Society Insulin Dependent Diabetes Trust JBOL King’s College Acute Trust La Leche League GB Leeds PCT Leeds Teaching Hospitals NHS Trust Liverpool PCT Liverpool Women’s Hospital NHS Trust vii .

Suffolk & Cambridgeshire Strategic Health Authority North Eastern Derbyshire PCT North Middlesex Hospital University Trust North Sheffield PCT North Tees and Hartlepool NHS Trust North Yorkshire and York PCT Northumbria Diabetes Service Northumbria Healthcare NHS Foundation Trust Northwest London Hospitals NHS Trust Nottingham City PCT Novo Nordisk Nutrition Society Obstetric Anaesthetists Association Pennine Acute Hospitals NHS Trust PERIGON (formerly the NHS Modernisation Agency) Pfizer Powys Local Health Board Primary Care Diabetes Society Primary Care Pharmacists Association PRIMIS+ Princess Alexandra Hospital NHS Trust Prodigy Queen Elizabeth Hospital NHS Trust Queen Mary’s Hospital NHS Trust (Sidcup) RCM Consultant Midwives Forum Regional Maternity Survey Office Roche Diagnostics Rotherham PCT Royal Bolton Hospitals NHS Trust Royal College of Anaesthetists Royal College of General Practitioners Royal College of General Practitioners Wales Royal College of Midwives Royal College of Nursing (RCN) Royal College of Obstetricians and Gynaecologists Royal College of Ophthalmologists viii .Diabetes in pregnancy Long Term Medical Conditions Alliance Luton and Dunstable Hospital NHS Trust Maidstone and Tunbridge Wells NHS Trust Maidstone Hospital Medicines and Healthcare products Regulatory Agency (MHRA) Medtronic Medway NHS Trust Menarini Diagnostics Merck Pharmaceuticals Mid and West Regional MSLC Mid Staffordshire General Hospitals NHS Trust MIDIRS (Midwives Information & Resource Service) Milton Keynes PCT National Audit Office National Childbirth Trust National Council for Disabled People and Carers from Black & Minority Ethnic Communities National Diabetes Consultant Nurse Group National Kidney Research Fund National Perinatal Epidemiology Unit National Public Health Service – Wales National Screening Committee Newcastle PCT NHS Clinical Knowledge Summaries Service NHS Direct NHS Quality Improvement Scotland Norfolk.

Guideline Development Group membership and acknowledgements Royal College of Paediatrics and Child Health Royal College of Pathologists Royal College of Physicians of Edinburgh Royal College of Physicians of London Royal College of Radiologists Royal Cornwall Hospitals NHS Trust and Peninsula Medical School Royal Society of Medicine Royal United Hospital Bath NHS Trust Royal West Sussex Trust Salford PCT Salford Royal Hospitals Foundation NHS Trust Sandwell and West Birmingham NHS Trust Sandwell PCT Sanofi-Aventis Scarborough and North Yorkshire Healthcare NHS Trust School of Midwifery Scottish Executive Health Department Scottish Intercollegiate Guidelines Network (SIGN) Scottish Nutrition & Diet Resources Initiative Sheffield PCT Sheffield Teaching Hospitals NHS Trust Society & College of Radiographers South Asian Health Foundation South Staffordshire PCT South Tees Hospitals NHS Trust Southend Hospitals NHS Trust Staffordshire Moorlands PCT Stockport PCT Sunderland Royal Hospital NHS Trust Sure Start Tamworth Syner-Med Pharmaceutical Products Takeda UK Tameside Acute Trust Tameside and Glossop Acute Services NHS Trust Trafford PCT UK Anaemia UK Clinical Pharmacy Association UK National Screening Committee UK Specialised Services Public Health Network University College London Hospitals NHS Trust University Hospital Lewisham University Hospital of North Staffordshire Acute Trust University Hospitals of Leicester University of Leicester (The Infant Mortality & Morbidity Studies) Welsh Assembly Government (formerly National Assembly for Wales) Welsh Endocrine and Diabetes Society Welsh Scientific Advisory Committee (WSAC) West Hertfordshire Hospitals Trust West Herts PCT West Middlesex University Hospital NHS Trust West Midlands Antenatal Diabetes Association Western Cheshire Primary Care Trust Whittington Hospital Trust Wiltshire PCT Wirral Hospital NHS Trust Worcestershire Acute Trust Worthing Hospital York NHS Trust Yorkshire and the Humber LSA ix .

Abbreviations AC ACE ACHOIS ADA AFP AGA ARB BMI BP CBG CEMACH CI CINAHL CSII DAFNE DCCT DESMOND DKA DPP DVLA EFW eGFR EL EPO ETDRS EUROCAT FBG FCG FPG GCT GDG GI GP HAPO HbA1c hCG HTA ICER IGT IUGR LGA LR MIG MDI MODY MoM NCC-WCH NDDG NHS NHS EED NHSLA NICE x abdominal circumference angiotensin-converting enzyme Australian Carbohydrate Intolerance Study in Pregnant Women American Diabetes Association alpha fetoprotein appropriate for gestational age angiotensin-II receptor blocker (also known as angiotensin-II receptor antagonist) body mass index blood pressure capillary blood glucose Confidential Enquiry into Maternal and Child Health confidence interval Cumulative Index to Nursing and Allied Health Literature continuous subcutaneous insulin infusion Dose Adjustment for Normal Eating Diabetes Control and Complications Trial Diabetes Education and Self Management for Ongoing and Newly Diagnosed diabetic ketoacidosis Diabetes Prevention Program Driver and Vehicle Licensing Agency estimated fetal weight estimated glomerular filtration rate evidence level erythropoietin Early Treatment Diabetic Retinopathy Study European Surveillance of Congenital Anomalies fasting blood glucose fasting capillary glucose fasting plasma glucose glucose challenge test guideline development group glycaemic index general practitioner Hyperglycemia and Adverse Pregnancy Outcome glycosylated haemoglobin human chorionic gonadotropin health technology assessment incremental cost-effectiveness ratio impaired glucose tolerance intrauterine growth restriction large for gestational age likelihood ratio Metformin in Gestational Diabetes multiple daily injection maturity-onset diabetes of the young multiple-of-median National Collaborating Centre for Women’s and Children’s Health National Diabetes Data Group National Health Service NHS Economic Evaluation Database National Health Service Litigation Authority National Institute for Health and Clinical Excellence .

Abbreviations NICU NNH NNT NPDR NPV NSF NT OGTT ONS OR PAPP-A PDR PPIP PPV QALY RBG RCT ROC RR SD SE SGA SPC TA TGA uE3 VBAC WHO neonatal intensive care unit number needed to harm number needed to treat non-proliferative diabetic retinopathy negative predictive value National Service Framework nuchal translucency oral glucose tolerance test Office for National Statistics odds ratio pregnancy-associated plasma protein-A proliferative diabetic retinopathy Patient and Public Involvement Programme positive predictive value quality-adjusted life year random blood glucose randomised controlled trial receiver operating characteristic relative risk standard deviation standard error small for gestational age summary of product characteristics technology appraisal transposition of the great arteries unconjugated estriol vaginal birth after previous caesarean section World Health Organization xi .

All subjects are then assessed with respect to things that happened to them in the past. The period of time in pregnancy preceding birth. For examples see selection bias. environmental factors) had been ruled out.Glossary of terms ACE inhibitor Angiotensin-converting enzyme inhibitor.g. analysis. The strongest research evidence available to support a particular guideline recommendation. Bias can occur at different stages in the research process. there is a causal relationship between a treatment and a disease if it can be shown that the treatment changes the course or outcome of the disease. Usually randomised controlled trials are needed to ascertain causality. Formal assessment of the quality of research evidence and its relevance to the clinical question or guideline under consideration. Influences on a study that can lead to invalid conclusions about a treatment or intervention. For example. The fetal heart rate is recorded by means of either an external ultrasonic abdominal transducer or a fetal scalp electrode. See also double-blind study. people without the disease). Bias can occur by chance or as a result of systematic errors in the design and execution of a study. Bias can even make it look as if the treatment works when it actually does not. The purpose of ‘blinding’ or ‘masking’ is to protect against bias. triple-blind study. performance bias. Detailed report on one patient (or case). things that might be related to getting the disease under investigation. The extent to which the results of a study or review can be applied to the target population for a clinical guideline. indicating renal dysfunction. Graphical representation of electronic monitoring of the fetal heart rate and of uterine contractions.g. The study starts with the identification of a group of individuals sharing the same characteristics (e. Proving cause and effect is much more difficult than just showing an association between two variables. a clinical trial in which the participating patients or their doctors are unaware of whether they (the patients) are taking the experimental drug or a placebo (dummy treatment). and everyone who avoided that food remained well. used in the assessment of obesity. For example. There is no comparison (control) group of patients. Description of several cases of a given disease. if it happened that everyone who had eaten a particular food became sick. The practice of keeping the investigators or subjects of a study ignorant of the group to which a subject has been assigned. it could not be proved that the food caused the sickness – unless all other possible causes (e. usually covering the course of the disease and the response to treatment. e. usually covering the course of that person’s disease and their response to treatment. publication bias. e. interpretation. people with a particular disease) and a suitable comparison (control) group (e. publication or review of research data.g. according to predetermined criteria. Enlargement of the heart. in the collection. Uterine contractions are recorded by means of an abdominal pressure transducer. Describes the relationship between two variables whenever it can be established that one causes the other. single-blind study. a class of drugs that reduce peripheral arterial resistance by inactivating an enzyme that converts angiotensin-I to the vasoconstrictor angiotensin-II. A study that compares exposure in subjects who have a particular outcome with those who do not. The presence of albumin in the urine. The body’s weight in kilograms divided by the square of the height in metres. information bias. confounding factor. even if leftovers were found to be contaminated.g. Such studies are also called retrospective as they look back in time from the outcome to the possible causes. For example. Bias in research can make a treatment look better or worse than it really is.g. then the food would clearly be associated with the sickness. However. Albuminuria Antenatal Applicability Appraisal of evidence Best available evidence Bias Blinding or masking Body mass index (BMI) Cardiomegaly Cardiotocograph Case report (or case study) Case series Case–control study Causal relationship xii .

It is usual to interpret a ‘95%’ confidence interval as the range of effects within which we are 95% confident that the true effect lies. doctor. Within the cycle there are stages that follow a systematic process of establishing best practice. Whereas ‘guidelines’ define what the best clinical practice should be. Cohorts can be assembled in the present and followed into the future (a ‘concurrent’ or ‘prospective’ cohort study) or identified from past records and followed forward from that time up to the present (a ‘historical’ or ‘retrospective’ cohort study). e. followed up in a research study for a specified period of time. it is called a focused question.g. An observational study that takes a group (cohort) of patients and follows their progress over time in order to measure outcomes such as disease or mortality rates and make comparisons according to the treatments or interventions that patients received. Each trial is designed to answer scientific questions and to find better ways to treat individuals with a specific disease. Coexistence of a disease or diseases in the people being studied in addition to the health problem that is the subject of the study. A way of expressing certainty about the findings from a study or group of studies. This general term encompasses controlled clinical trials and randomised controlled trials. The Cochrane Library is available on CD-ROM and the internet. Thus within the study group. A wide confidence interval indicates a lack of certainty or precision about the true size of the clinical effect and is seen in studies with too few patients. An international organisation in which people find. Clinical audit can be described as a cycle or spiral. The importance of a particular guideline recommendation to the clinical management of the target population. measuring care against specific criteria. Where confidence intervals are narrow they indicate more precise estimates of effects and a larger sample of patients studied. e. patients with the same disease). (Clinical trials that assess effectiveness are sometimes called management trials. these subgroups may be quite different in their characteristics and some adjustment must be made when analysing the results to ensure that the comparison between groups is as fair as possible. physiotherapist. subgroups of patients are identified (from information collected about patients) and these groups are compared with respect to outcome.g. ‘audit’ investigates whether best practice is being carried out. A research study conducted with patients which tests out a drug or other intervention to assess its effectiveness and safety. each cycle aspires to a higher level of quality. appraise and review specific types of studies called randomised controlled trials.Glossary of terms Centile Checklist Clinical audit Any of the 99 numbered points that divide an ordered set of scores into 100 parts each of which contains one-hundredth of the total. or treatment outcomes. The Cochrane Library consists of a regularly updated collection of evidence-based medicine databases including the Cochrane Database of Systematic Reviews (reviews of randomised controlled trials prepared by the Cochrane Collaboration). The Cochrane Database of Systematic Reviews contains regularly updated reviews on a variety of health issues and is available electronically as part of the Cochrane Library. A confidence interval describes a range of possible effects (of a treatment or intervention) that are consistent with the results of a study or group of studies. When a clinical question is formulated in a precise way. has a beneficial effect on the course or outcome of disease compared with no treatment or other routine care. and monitoring to sustain improvement. See study checklist. when used under usual or everyday conditions. of the target population. The spiral suggests that as the process continues. This term is sometimes used in guideline development work to refer to the questions about treatment and care that are formulated in order to guide the search for research evidence. Clinical effectiveness Clinical impact Clinical importance Clinical question Clinical trial Clinician Cochrane Collaboration Cochrane Library Cohort Cohort study Comorbidity Confidence interval (CI) xiii . A systematic process for setting and monitoring standards of clinical care. comparing mortality between one group that received a specific treatment and one group which did not (or between two groups that received different levels of treatment). A group of people sharing some common characteristic (e. taking action to improve care. nurse. The extent to which a specific treatment or intervention.g. Because patients are not randomly allocated to subgroups. The effect that a guideline recommendation is likely to have on the treatment. A qualified healthcare professional providing patient care. using statistical techniques.) Clinical ‘effectiveness’ is not the same as efficacy.

A type of economic evaluation where both costs and benefits of healthcare treatment are measured in the same monetary units. The observation of a defined set of people at a single point in time or time period – a snapshot. A treatment is assessed in terms of its ability to both extend life and to improve the quality of life. A statement of the advised course of action in relation to a particular clinical topic. This type of study contrasts with a longitudinal study. A. A group of patients recruited into a study that receives no treatment. A study comparing two or more interventions in which the participants. half the participants are randomly allocated to receive them in the order A. The group then retires to consider the questions in the light of the evidence presented and attempts to reach a consensus. to assess its applicability to the target population and the strength of any recommendation that it would support. Formal consensus methods include Delphi and nominal group techniques and consensus development conferences. If benefits exceed costs. the cost of preventing one additional heart attack. if a group of people exercising regularly and a group of people who do not exercise have an important age difference then any difference found in outcomes about heart disease could well be due to one group being older than the other rather than due to the exercising. A physical or biochemical malformation which is present at birth A technique used for the purpose of reaching an agreement on a particular issue. The extent to which the conclusions of a collection of studies used to support a guideline recommendation are in agreement with each other. A specific healthcare treatment is said to be ‘cost-effective’ if it gives a greater health gain than could be achieved by using the resources in other ways. such as a new drug. which means allowing sufficient time between stopping one treatment and starting another so that the first treatment has time to wash out of the patient’s system. consensus methods may be used where there is a lack of strong research evidence on a particular topic. A type of economic evaluation comparing the costs and the effects on health of different treatments. One (the experimental group) receives the treatment that is being tested. Congenital anomaly Consensus development conference Consensus methods Consensus statement Considered judgement Consistency Control group Controlled clinical trial (CCT) Cost–benefit analysis Cost-effectiveness Cost-effectiveness analysis Cost–utility analysis Crossover study design Cross-sectional study Data set xiv . A special form of cost-effectiveness analysis where health effects are measured in qualityadjusted life years. For example. upon completion of the course of one treatment. Health effects are measured in ‘health-related units’. based on the collective views of a body of experts. Therefore a crossover study should include an adequate ‘wash-out’ period. for example. a treatment of known effect or a placebo (dummy treatment) in order to provide a comparison for a group receiving an experimental treatment. a placebo (dummy treatment) or no treatment. See also consensus methods. A list of required information relating to a specific disease. Value for money. and the other (the comparison or control group) receives an alternative treatment. A CCT where patients are randomly allocated to treatment and comparison groups is called a randomised controlled trial. For example. which follows a set of people over a period of time. It influences a study and can contribute to misleading findings if it is not understood or appropriately dealt with. It involves bringing together a group of about ten people who are presented with evidence by various interest groups or experts who are not part of the decision-making group. A problem with this study design is that the effects of the first treatment may carry over into the period when the second is given. A study testing a specific drug or other treatment involving two (or more) groups of patients with the same disease. Age is the confounding factor here and the effect of exercising on heart disease cannot be assessed without adjusting for age differences in some way. the evaluation would recommend providing the treatment. A variety of techniques that aim to reach an agreement on a particular issue. are switched to another.Diabetes in pregnancy Confounder or confounding factor A factor that can bring an alternative explanation to an association observed between an exposure and the outcome of interest. for a comparison of treatments A and B. In the development of clinical guidelines. The two groups are followed up to compare differences in outcomes to see how effective the experimental treatment was. See also homogeneity. B and half to receive them in the order B. The application of the collective knowledge of a guideline development group to a body of evidence.

in a laboratory) has a beneficial effect on the course or outcome of disease compared with no treatment or other routine care. A process by which members of a working group or committee ‘declare’ any personal or professional involvement with a company (or related to a technology) that might affect their objectivity. costs and health consequences associated with each option. It consists of the probabilities. The extent to which a specific treatment or intervention. A state of absolute or relative insulin deficiency characterised by hyperglycaemia. participants are asked to give further views in the light of the group feedback. under ideally controlled conditions (e. A technique used for the purpose of reaching an agreement on a particular issue. A comparison of alternative courses of action in terms of both their costs and consequences. After the first questionnaire. The less effective and more costly option is said to be ‘dominated’. Also called level of evidence.Glossary of terms Decision analysis The study of how people make decisions or how they should make decisions. A study to assess the effectiveness of a test or measurement in terms of its ability to accurately detect or exclude a specific disease. The overall effectiveness or overall cost-effectiveness of different actions can then be compared. The purpose of blinding is to protect against bias. A code (e. A complication of diabetes affecting the blood vessels in the retina at the back of the eye. There may be bleeding from retinal vessels (non-proliferative retinopathy) or the development of new abnormal vessels (proliferative retinopathy). There are several methods that decision analysts use to help people to make better decisions. dehydration. The process of systematically finding. A method for helping people to make better decisions in situations of uncertainty. represent the evidence supporting a particular recommendation or series of recommendations in a guideline. See also consensus methods. See clinical effectiveness. without the participants meeting or interacting directly. xv . for example if their position or department is funded by a pharmaceutical company. 1+) linked to an individual study. A table summarising the results of a collection of studies which. Clinical procedures that are regarded as advantageous to the patient but not urgent. appraising and using research findings as the basis for clinical decisions. acidosis and ketosis. taken together. A term used in health economics describing when an option for treatment is both less clinically effective and more costly than an alternative option. Based directly on experience (observation or experiment) rather than on reasoning alone. It illustrates the decision as a succession of possible actions and outcomes. 1++. The judgements of the participants are statistically aggregated.g. Decision tree Declaration of interest Delphi method Diabetic ketoacidosis (DKA) Diabetic nephropathy Diabetic retinopathy Diagnostic study Dominance Double-blind study Economic evaluation Effectiveness Efficacy Elective Empirical Epidemiology Erb palsy Evidence based Evidence level Evidence table Evidence-based clinical practice Evidence-based clinical practice involves making decisions about the care of individual patients based on the best research evidence available rather than basing decisions on personal opinions or common practice (which may not always be evidence based). Injury to the nerve roots of the brachial plexus of an arm mainly related to birth trauma and leading to various degrees of weakness of the affected arm which may resolve during the first year of life. sometimes after weighting for expertise. Study of diseases within a population. covering the causes and means of prevention. It involves sending participants a series of postal questionnaires asking them to record their views. In health economic evaluations the consequences should include health outcomes. including decision trees.g. Evidence-based clinical practice therefore involves integrating individual clinical expertise and patient preferences with the best available evidence from research. which can affect vision. indicating where it fits into the hierarchy of evidence and how well it has adhered to recognised research principles. A study in which neither the subject (patient) nor the observer (investigator/clinician) is aware of which treatment or intervention the subject is receiving. Kidney dysfunction or disease occurring as a result of diabetes.

Experimental treatment Extrapolation Fetal growth restriction Fetal surveillance Focused question Folic acid Funnel plot Gestation Gestational diabetes Glucose electrode Glycaemic control targets Glycosylated haemoglobin (HbA1c) Gold standard Grey literature Guideline Guideline recommendation Health economics Health technology Heterogeneity xvi . biophysical profiles and fetal heart monitoring. definition of variables or duration of follow-up. The process of performing fetal wellbeing tests. either estimated weight or abdominal circumference below the 10th percentile.Diabetes in pregnancy Exclusion criteria Experimental study See selection criteria. It is used to assist clinician and patient decision making about appropriate health care for specific clinical conditions. Or lack of homogeneity. A test which measures the amount of glucose-bound haemoglobin and reflects how well the blood glucose level has been controlled over the previous 2–3 months. Publication bias may lead to asymmetry in funnel plots. Blood glucose measurement using electrochemical biosensors. procedure or measurement that is widely accepted as being the best available.g. health promotion activities (e. They show the treatment effects estimated from separate studies on the horizontal axis against a measure of sample size on the vertical axis. fetal and placental Doppler ultrasounds. The time from conception to birth. and any comparisons that are to be made. surgical procedures. Reports that are unpublished or have limited distribution and are not included in bibliographic retrieval systems. rather than opinion. or slowing growth velocity of the abdominal circumference as measured at a subsequent ultrasound scan. These may include ultrasound scans. Carbohydrate intolerance of varying severity which is diagnosed in pregnancy and may or may not resolve after pregnancy. the treatment or intervention to be investigated. Questions are normally expected to identify the patients or population involved. the role of diet versus medicines in disease management) and other therapeutic interventions. Course of action advised by the guideline development group on the basis of their assessment of the supporting evidence. A research study designed to test if a treatment or intervention has an effect on the course or outcome of a condition or disease – where the conditions of testing are to some extent under the control of the investigator. A method. outcome measures. what outcomes are to be considered. Funnel plots are simple scatter plots on a graph. For example. a new drug) being studied to see if it has an effect on the course or outcome of a condition or disease. The duration of gestation is measured from the first day of the last normal menstrual period. do insulin pumps (intervention) improve blood sugar control (outcome) in adolescents with type 1 diabetes (population) compared with multiple insulin injections (comparison)? See also clinical question. A good guideline makes recommendations about treatment and care based on the best research available. Evidence of abnormally slow growth of the fetus within the uterus. Such results may occur as a result of differences between studies in terms of the patient populations. medical devices such as artificial hip joints. A study question that clearly identifies all aspects of the topic that are to be considered while seeking an answer. diagnostic techniques. Controlled clinical trial and randomised controlled trial are examples of experimental studies.g. The application of research evidence based on studies of a specific population to another population with similar characteristics. Recommended levels of blood glucose. A treatment or intervention (e. A systematically developed tool that describes aspects of a patient’s condition and the care to be given. A water-soluble vitamin in the B-complex group which helps to prevent fetal neural tube defect when commenced by the mother before conception. A branch of economics that studies decisions about the use and distribution of healthcare resources. Health technologies include medicines. The term is used in meta-analyses and systematic reviews when the results or estimates of effects of treatment from separate studies seem to be very different in terms of the size of treatment effects or even to the extent that some indicate beneficial and others suggest adverse treatment effects.

The overall approach of a research project.g.Glossary of terms Hierarchy of evidence An established hierarchy of study types. e. The spread of a set of values between which 25% (25th centile) and 75% (75th centile) of these values lie. The extent to which a study has conformed to recognised good practice in the design and execution of its research methods. Often defined as having a birthweight above the 90th centile for gestation or a birthweight of 4000 g or more.g. psychological therapy. it may be inappropriate or even misleading to statistically pool results in this way. etc. Low blood glucose level. Homogeneity Hypertension Hypoglycaemia Inclusion criteria Information bias Interquartile range (IQR) Intervention Level of evidence Literature review Longitudinal study Macrosomia Masking Median Meta-analysis Metformin Methodological quality Methodology Microalbuminuria Miscarriage Maturity-onset diabetes of the young (MODY) Multicentre study Multidisciplinary clinic xvii . observer or interviewer errors (e. lack of blinding). Oversized baby as seen. a faulty machine). drug treatment. difficult or biased questions). because of differences in the study populations or in the outcomes measured.g. See evidence level. several large statistically significant RCTs which are in agreement represent stronger evidence than one small RCT). Healthcare action intended to benefit the patient.g. See selection criteria. A very small increase in urinary albumin. e. A group of autosomal dominant disorders in young people each caused by a single gene defect. For diabetes. Results are usually regarded as homogeneous when differences between studies could reasonably be expected to occur by chance. reading and assessing the quality of published (and unpublished) articles on a given topic. This type of study contrasts with a cross-sectional study. See blinding. A clinic with access to care from health professionals in more than one discipline. Wellconducted randomised controlled trials (RCTs) are at the top of this hierarchy (for example. Spontaneous ending of a pregnancy before viability (currently taken as 24 weeks of gestation). Results from a collection of independent studies (investigating the same treatment) are pooled. diabetology.g. See also consistency. based on the degree of certainty that can be attributed to the conclusions that can be drawn from a well-conducted study. A study of the same group of people at more than one point in time. Pertinent to all types of study and can be caused by inadequate questionnaires (e. Well-conducted studies of patients’ views and experiences would appear at a lower level in the hierarchy of evidence. See also systematic review and heterogeneity. response errors (e. using statistical techniques to synthesise their findings into a single estimate of a treatment effect. the disciplines recommended are obstetrics. nursing. High blood pressure. This means that the results of studies included in a systematic review or meta-analysis are similar and there is no evidence of heterogeneity. associated with decreased insulin production and varying degrees of clinical severity. lack of blinding if patients are aware of the treatment they receive) and measurement error (e. which observes a defined set of people at a single point in time. surgical procedure. Where studies are not compatible e.g.g. A study where subjects were selected from different locations or populations. a cooperative study between different hospitals or an international collaboration involving patients from more than one country. A process of collecting. The value of the middle item of a series when the items are arranged in numerical order. that the study will be a randomised controlled trial of 200 people over 1 year. for example as a consequence of the effect of diabetes during pregnancy. An oral antidiabetic agent that decreases glucose production by the liver and lowers plasma glucose levels. for example. midwifery and dietetics.

g. A technique used for the purpose of reaching an agreement on a particular issue. if the NNH = 4. Negative predictive value (NPV) The proportion of people with a negative test result who do not have the disease (where not having the disease is indicated by the ‘gold standard’ test being negative). risk ratio.g. In research about diseases or treatments. There are different levels of complexity of care which can be offered by an individual neonatal unit. See also relative risk. the result is seen as highly significant. symptoms or situation of a person.g. The settings are usually natural. the change in health. if the NNT = 4. In other words. There is a greater risk of selection bias than in experimental studies. which can be used to measure the effectiveness of care/treatment/rehabilitation. We would conclude that there probably is a difference between treatments. functional ability. e. which is the number of patients that would need to receive a treatment to cause one additional adverse event. In recent years odds ratios have become widely used in reports of clinical studies.g.03. P values just tell us whether an effect can be regarded as statistically significant or not. without the intervention of the investigator. See also consensus methods. e. Researchers should decide what outcomes to measure before a study begins. outcomes are then assessed at the end of the study. A major birth defect caused by abnormal development of the neural tube. By convention. The end result of care and treatment and/or rehabilitation. for which we need the confidence interval. but they can be laboratory settings. They provide an estimate (usually with a confidence interval) for the effect of a treatment. Where the value of P is 0. In no way do they relate to how big the effect might be.) Suppose the P value was P = 0. less than 5%) the result is seen as statistically significant. Observational study Odds ratio (OR) Outcome P value xviii . Analogous to the NNT is the number needed to harm (NNH). with individual judgements being aggregated statistically to derive the group judgement. then four patients would have to be treated for one bad outcome to occur. this refers to a study in which nature is allowed to take its course. For rare events the odds ratio and the relative risk (which uses actual risks and not odds) will be very similar. Odds are a way of representing probability.e. An odds ratio of greater than 1 shows an increased risk and less than 1 shows a protective effect. It states how many patients need to be treated with the treatment in question in order to prevent an event which would otherwise occur. especially familiar for betting. See review. (The assumption that there really is no difference between treatments is called the ‘null hypothesis’. whether or not people received a specific treatment or intervention) are studied in relation to changes or differences in other(s) (e. If a study is done to compare two treatments then the P value is the probability of obtaining the results of that study or something more extreme if there really was no difference between treatments. Neonatal death Neonatal unit Death of a liveborn baby before 28 completed days after birth. A study based on subjects selected on the basis of their availability. Changes or differences in one characteristic (e. It uses a variety of postal and direct contact techniques. defined as a body mass index of 30 kg/m² or greater.001 or less. listening to and recording behaviours.Diabetes in pregnancy Multiparous A woman who has had at least one previous birth (from 24 weeks onwards). Odds are used to convey the idea of ‘risk’ and an odds ratio of 1 between two treatment groups would imply that the risks of an adverse outcome were the same in each group. as in psychological research. It involves watching. Neural tube defect Nominal group technique Non-experimental study Non-systematic review Number needed to treat (NNT) This measures the impact of a treatment or intervention. where the value of P is below 0. What this means is that if there really was no difference between treatments then there would only be a 3% chance of getting the kind of results obtained. whether or not they died). then four patients would have to be treated to prevent one bad outcome. the better the treatment is. actions. Obesity Observation Increased body weight. Since this chance seems quite low we should question the validity of the assumption that there really is no difference between treatments. the structure present during embryonic life which later gives rise to the central nervous system (brain and spinal cord). They are a measure of the excess risk or degree of protection given by exposure to a certain factor.05 (i. A research technique used to help understand complex situations. activities and interactions. A unit which provides additional care for babies over and above that which can be offered on a postnatal ward or transitional care unit. with no attempt having been made to avoid problems of bias. The closer the NNT is to 1.

A small scale ‘test’ of the research instrument. Protocol Publication bias Quality-adjusted life years (QALYs) Quantitative research Quintile Random allocation or randomisation xix . Systematic differences in care provided apart from the intervention being evaluated. and care providers may treat patients differently according to what group they are in. what question is to be answered and how information will be collected and analysed. Random allocation implies that each individual (or each unit in the case of cluster randomisation) being entered into a study has the same chance of receiving each of the possible interventions. how likely a treatment or intervention will alleviate a symptom. QALYs are calculated by estimating the years of life remaining for a patient following a particular care pathway and weighting each year with a quality of life score (on a zero to one scale). Masking (blinding) of both the recipients and providers of care is used to protect against performance bias.Glossary of terms Peer review Review of a study. Meta-analyses that are exclusively based on published literature may therefore produce biased results. See statistical power. One QALY is equal to 1 year of life in perfect health or 2 years at 50% health. Research that generates numerical data or data that can be converted into numbers. by using a random numbers table or a computer-generated random sequence. Studies with statistically significant results are more likely to get published than those with non-significant results. e. How likely an event is to occur.g. for example testing out (piloting) a new questionnaire with people who are similar to the population of the study in order to highlight any problems or areas of concern. A plan or set of steps that defines appropriate action. The portion of a frequency distribution containing one-fifth of the total sample. and so on. This type of bias can be assessed by a funnel plot. Peer reviewers can include professional and/or patient/carer representatives. in advance of carrying out the study. for example clinical trials or the national Census that counts people and households. Birth before 37 weeks and 0 days of gestation. which can then be addressed before the full-scale study begins. A study in which people are entered into the research and then followed up over a period of time with future events recorded as they happen. both at national and local levels. A research protocol sets out. A beneficial (or adverse) effect produced by a placebo and not due to any property of the placebo itself. Performance bias Pilot study Placebo Placebo effect Positive predictive value (PPV) The proportion of people with a positive test result who have the disease (where having the disease is indicated by the ‘gold standard’ test being positive). Postnatal Power Preterm birth Prevalence Probability Prospective study The period of time occurring after birth. people who know they are in the experimental group may experience placebo effects. Guideline implementation protocols set out how guideline recommendations will be used in practice by the NHS. Fake or inactive treatments received by participants allocated to the control group in a clinical trial that are indistinguishable from the active treatments being given in the experimental group. For example. This contrasts with studies that are retrospective. if study participants know they are in the control group they may be more likely to use other forms of care. for example. A measure of health outcome that looks at both length of life and quality of life. They are used so that participants are ignorant of their treatment allocation in order to be able to quantify the effect of the experimental treatment over and above any placebo effect due to receiving care or attention. service or recommendations by those with similar interests and expertise to the people who produced the study findings or recommendations. A method that uses the play of chance to assign participants to comparison groups in a research study. The proportion of individuals in a population having a disease.

an adverse reaction to the drug being tested) in one group of subjects compared with another group. With physical assessments it is possible for different clinicians to make independent assessments in quick succession and if their assessments tend to agree then the method of assessment is said to be reliable. a relative risk of 2 would indicate that patients receiving one of the treatments had twice the risk of an undesirable outcome than those receiving the other treatment. Explicit standards used by guideline development groups to decide which studies should be included and excluded from consideration as potential sources of evidence.Diabetes in pregnancy Randomised controlled trial (RCT) A study to test a specific drug or other treatment in which people are randomly assigned to two (or more) groups: one (the experimental group) receiving the treatment that is being tested. The term relative risk is sometimes used as a synonym of risk ratio. Selection bias has occurred if: • the characteristics of the sample differ from those of the wider population from which the sample has been drawn. Examining the fundi through pupils which have been dilated with eye drops. Range Relative risk (RR) Reliability Retinal assessment Retrospective study Review Risk ratio Sample Sampling Scottish Intercollegiate Guidelines Network (SIGN) Selection bias Selection criteria Semi-structured interview Sensitivity Severe hypoglycaemia Shoulder dystocia Single-blind study Singleton xx . For example. A study in which either the subject (patient/participant) or the observer (clinician/ investigator) is not aware of which treatment or intervention the subject is receiving. Through randomisation. Ratio of the risk of an undesirable event or outcome occurring in a group of patients receiving experimental treatment compared with a comparison (control) group. and the other (the comparison or control group) receiving an alternative treatment. 100% sensitivity means that all those with a negative test result do not have the disease. Specificity should be considered alongside sensitivity to fully judge the accuracy of a test. The two groups are followed up to compare differences in outcomes to see how effective the experimental treatment was. A part of the study’s target population from which the subjects of the study will be recruited. This contrasts with studies that are prospective. The way participants are selected for inclusion in a study. following up areas of interest in response to the information given by the respondent. In a study comparing two treatments. If subjects are drawn in an unbiased way from a particular population. A semi-structured interview allows more flexibility than a structured interview. someone who has a high score on one occasion tends to have a high score if measured on another occasion very soon afterwards. A summary measure which represents the ratio of the risk of a given event or outcome (e. When the ‘risk’ of the event is the same in the two groups the relative risk is 1. Any documented evidence of difficulty with delivering the shoulders after delivery of the baby’s head. The interviewer asks a number of open-ended questions. One fetus or baby. a placebo (dummy treatment) or no treatment. Structured interviews involve asking people pre-set questions. sensitivity is the proportion of true positive results that are correctly identified as positive by the test. In diagnostic testing. the results can be generalised from the sample to the population as a whole. or • there are systematic differences between comparison groups of patients in a study in terms of prognosis or responsiveness to treatment. Relative risk is sometimes used as a synonym for risk ratio. A review is considered non-systematic unless an extensive literature search has been carried out to ensure that all aspects of the topic are covered and an objective appraisal made of the quality of the studies. The difference or interval between the smallest and largest values in a frequency distribution. A study that deals with the present/past and does not involve studying future events.g. Hypoglycaemia requiring help from another person. the groups should be similar in all aspects apart from the treatment they receive during the study. Reliability refers to a method of measurement that consistently gives the same results. SIGN was established in 1993 to sponsor and support the development of evidencebased clinical guidelines for the NHS in Scotland. Summary of the main points and trends in the research literature on a specified topic.

disease state. 80% is an acceptable level of power. It accounts for 5–15% of all people with diabetes. 100% specificity means that all those with a positive test result have the disease.Glossary of terms Sliding scale Specificity Intravenous insulin and dextrose infusions with a set of instructions for adjusting the dose of insulin on the basis of blood glucose test results. due to autoimmune destruction of the insulin-producing beta cells in the islets of Langerhans in the pancreas. Sensitivity should be considered alongside specificity to fully judge the accuracy of a test. A study in which information is systematically collected from people (usually from a sample within a defined population).g. See also P value. A unit providing care of term or near-term babies not needing high-dependency or intensive care. case–control studies. By convention. social background. A study in which the statistical analysis is carried out without knowing which treatment patients received. In diagnostic testing. as undertaken by NICE. specificity is the proportion of true negative results that are correctly identified as negative by the test. One of the 3 month periods into which pregnancy is divided. A measure of the spread. for example in terms of age. A technology appraisal. a statistically significant treatment effect) if there really was an important difference (e. Usually used with the mean (average) to describe numerical data. A person who takes part in an experiment or research study. Randomised controlled trials. the second trimester is 14–26 weeks of gestation. May or may not include a meta-analysis. If the statistical power of a study is low. People who have been identified as the subjects of a study. The ability of a study to demonstrate an association or causal relationship between two variables. Standard deviation Statistical power Stillbirth Study checklist Study population Study quality Study type Subject Survey Systematic Systematic error Systematic review Target population Technology appraisal (TA) Thrombosis Transitional care unit Trimester Triple-blind study Type 1 diabetes xxi . in addition to the patients and investigators/clinicians being unaware which treatment patients were getting. appraised and synthesised in a methodical way according to predetermined criteria. The people to whom guideline recommendations are intended to apply. which can be safely delivered without babies being separated from their mothers. A list of questions addressing the key aspects of the research methodology that must be in place if a study is to be accepted as valid. and the third trimester is 27 weeks of gestation until birth. not random. The formation or presence of a clot of coagulated blood in a blood vessel. A review in which evidence from scientific studies has been identified. The first trimester is 0– 13 weeks of gestation. These checklists are used to ensure a degree of consistency in the way that studies are evaluated. Legal definition: a child that has issued forth from its mother after the 24th week of pregnancy and which did not at any time after being completely expelled from its mother breathe or show any other signs of life (Section 41 of the Births and Deaths Registration Act 1953 as amended by the Stillbirth Definition Act 1992). NICE technology appraisals are designed to provide patients. and cohort studies are all examples of study types. Methodical. scatter or variability of a set of measurements. There is an absolute deficiency of insulin production. 80% power in a clinical trial means that the study has a 80% chance of ending up with a P value of less than 5% in a statistical test (i. the study results will be questionable (the study might have been too small to detect any differences). given that an association exists. The kind of design used for a study. 10% versus 5% mortality) between treatments. is the process of determining the clinical and cost-effectiveness of a health technology. Recommendations may be less valid if applied to a population with different characteristics from the participants in the research study. Refers to the various errors or biases inherent in a study. health professionals and managers with an authoritative source of advice on new and existing health technologies. A different checklist is required for each study type. See methodological quality. See also bias.e. according to plan. For example.

A measurement that can vary within a study. It accounts for 85–95% of all people with diabetes. xxii .g. e. Variability is present when differences can be seen between different people or within the same person over time with respect to any characteristic or feature that can be assessed or measured.Diabetes in pregnancy Type 2 diabetes Variable There is a relative deficiency of insulin production. and/or the insulin produced is not effective (insulin resistance). the age of participants.

neuropathy and vascular disease. with 7. hypoglycaemia and ketoacidosis) and diabetic complications (such as retinopathy) during pregnancy the timetable of antenatal appointments to be offered to women with diabetes timing and mode of birth (including induction of labour. Approximately 87.2 There is a lack of data about the prevalence of gestational diabetes.1).5% of pregnancies complicated by diabetes are. respectively.4.5 Miscarriage. Approximately 650 000 women give birth in England and Wales each year. black Caribbean.10% of births. with certain minority ethnic groups being at increased risk.27% and 0.20 Pre-existing type 1 diabetes and pre-existing type 2 diabetes account for 0. antenatal and intrapartum periods changes to medications for diabetes and its complications before or during pregnancy management of diabetic emergencies (for example. 1 . nephropathy. In addition. It focuses on areas where additional or different care should be offered to women with diabetes and their newborn babies. which may or may not resolve after pregnancy. caesarean section. birth injury.1 and 2–5% of pregnancies involve women with diabetes.3.1 1. 1. Diabetes in pregnancy is associated with risks to the woman and to the developing fetus. In particular.5% being due to type 1 diabetes and the remaining 5% being due to type 2 diabetes. see Section 4. congenital malformations. In its chronic forms. Middle Eastern and Chinese family origin).6 This clinical guideline concerns the management of diabetes and its complications from preconception to the postnatal period. depending on factors such as ethnic origin. analgesia and anaesthesia.2. including retinopathy. macrosomia. and the use of steroids for fetal lung maturation) initial care of the newborn baby management of diabetes and its complications during the postnatal period. It has been developed with the aim of providing guidance on: • • • • • • • • • preconception information diagnosis and management of gestational diabetes glycaemic control in the preconception. type 2 diabetes is increasing in certain minority ethnic groups (including people of African. Stillbirth. pre-eclampsia and preterm labour are more common in women with pre-existing diabetes. This clinical guideline contains recommendations for the management of diabetes and its complications in women who wish to conceive and those who are already pregnant. South Asian. therefore.1 Introduction Diabetes in pregnancy Diabetes is a disorder of carbohydrate metabolism that requires immediate changes in lifestyle. perinatal mortality and postnatal adaptation problems (such as hypoglycaemia) are more common in babies born to women with pre-existing diabetes.5% (the precise figure varies from region to region.2 Aim of the guideline Clinical guidelines have been defined as ‘systematically developed statements which assist clinicians and patients in making decisions about appropriate treatment for specific conditions’. estimated to be due to gestational diabetes. diabetic retinopathy can worsen rapidly during pregnancy. The guideline builds on existing clinical guidelines for routine care during the antenatal.2 The prevalence of type 1 and type 2 diabetes is increasing. intrapartum and postnatal periods. diabetes is associated with long-term vascular complications. The clinical experience of the guideline development group (GDG) suggests that the average prevalence in England and Wales is approximately 3.

All GDG members’ and external advisers’ potential and actual conflicts of interest were recorded on declaration forms provided by NICE and are presented in Appendix A. A version of this guideline for women with diabetes and the public is available from the NICE website (www. trust and care-home managers • women with diabetes and their families. Health Commission Wales commissioners. and public health. diabetes physicians and neonatal paediatricians) • those responsible for commissioning and planning healthcare services. personal non-pecuniary interests (including research interests). nurses and midwives. fee-paid work. and included the following: • • • • • • • two obstetricians two diabetes physicians two diabetes specialist midwives a diabetes specialist nurse a GP a neonatal paediatrician two patient/carer representatives. Staff from the NCC-WCH provided methodological support for the guideline development process by undertaking systematic searches. health economic modelling and writing successive drafts of the guideline. Expert advisers were appointed by the GDG to advise on each of these issues. intrapartum and postnatal care that apply equally to women with or without diabetes • aspects of routine care for women with diabetes that do not change during the preconception. The membership of the GDG was determined by the NCC-WCH and NICE. personal family interests (including shareholdings) and non-personal pecuniary 2 . although they were not involved in the final decisions regarding formulation of recommendations. which covers England. in particular: • healthcare professionals involved in the care of women with diabetes and their newborn babies (including general practitioners (GPs). The forms covered personal pecuniary interests (including consultancies.uk/CG063publicinfo) or from NICE publications on 0845 003 7783 (quote reference number N1485). antenatal.Diabetes in pregnancy 1.nice. retrieving and appraising the evidence. shareholdings. The neonatal paediatrician appointed to the GDG at the beginning of the development process resigned in October 2007 due to ill health and was replaced by another neonatal paediatrician.3 Areas outside the remit of the guideline This guideline does not address: • aspects of routine antenatal. Wales and Northern Ireland. obstetricians. including primary care trust commissioners. During the development of the guideline.org.4 For whom is the guideline intended? This guideline is of relevance to those who work in or use the National Health Service (NHS) in England. fellowships and support from the healthcare industry). the GDG identified a need for expert advice in relation to obstetric analgesia and anaesthesia. diabetic retinopathy and data relating to pregnancy in women with pre-existing diabetes held by the Confidential Enquiry into Maternal and Child Health (CEMACH). 1. intrapartum or postnatal periods • advice about contraceptive methods for women with diabetes • management of morbidity in newborn babies of women with diabetes beyond initial assessment and diagnosis. Wales and Northern Ireland.5 Who has developed the guideline? The National Collaborating Centre for Women’s and Children’s Health (NCC-WCH) was commissioned by the National Institute for Health and Clinical Excellence (NICE) to establish a multi-professional and lay working group (the GDG) to develop the guideline. 1.

London: NICE. See National Institute for Health and Clinical Excellence. 2nd ed. 2008.20  The type 2 diabetes guideline has been published since the diabetes in pregnancy guideline was first printed. renal disease and retinopathy) – Antenatal care: routine care for the healthy pregnant woman9 (the section of the antenatal care guideline that addresses screening for gestational diabetes has been developed in parallel with the development of this guideline. The different types of organisations that were eligible to register as stakeholders included: • national patient and carer organisations that directly or indirectly represent the interests of women with diabetes and their families before and during pregnancy • national organisations that represent the healthcare professionals who provide services for women with diabetes before and during pregnancy • companies that manufacture the preparations or products used in the management of diabetes in pregnancy • providers and commissioners of health services in England.  The information on type 2 diabetes in this TA has been updated since the diabetes in pregnancy guideline was first printed by the publication of National Institute for Health and Clinical Excellence.13 • Technology appraisals (TAs) – Guidance on the use of continuous subcutaneous insulin infusion for diabetes14  – Guidance on the use of glitazones for the treatment of type 2 diabetes15 – Guidance on the use of long-acting insulin analogues for the treatment of diabetes – insulin glargine17 – Guidance on the use of patient-education models for diabetes. 2008. Type 2 Diabetes: the Management of Type 2 Diabetes. – Intrapartum care: care of healthy women and their babies during childbirth10 – Postnatal care: routine postnatal care of women and their babies11 – Induction of labour12 (this diabetes in pregnancy guideline replaces the part of the induction of labour guideline that relates to women with diabetes) – Caesarean section. 2008.  The guidance on the use of continuous subcutaneous insulin infusion for diabetes has been updated since the diabetes in pregnancy guideline was first printed. Registered stakeholders were consulted throughout the guideline development process. young people and adults7 – Type 2 diabetes: the management of type 2 diabetes8 (to replace separate guidelines on blood glucose.Introduction interests (including funding from the healthcare industry for research projects and meetings).  The induction of labour guideline has been updated since the diabetes in pregnancy guideline was first printed. The process of stakeholder registration was managed by NICE. Type 2 Diabetes: the Management of Type 2 Diabetes. London: NICE. London: RCOG Press. See National Collaborating Centre for Women’s and Children’s Health. See National Institute for Health and Clinical Excellence. including the following NICE guidance. 2008.18  • Public health guidance – Improving the nutrition of pregnant and breastfeeding mothers and children in lowincome households.19 The guideline also complements and updates the National Service Framework (NSF) for diabetes. The GDG chair and NCC-WCH project director considered all the declarations and concluded that none of the declared interests constituted a material conflict of interest that would influence the recommendations developed by the GDG.6 Other relevant documents This guideline is intended to complement other existing and proposed works of relevance. blood pressure and blood lipids. Induction of Labour (2008 update). Organisations with interests in the management of diabetes and its complications from preconception to the postnatal period were encouraged to register as stakeholders for the guideline. 3 . 1. Continuous Subcutaneous Insulin Infusion for Treatment of Diabetes Mellitus. Wales and Northern Ireland • statutory organisations such as the Department of Health and the Welsh Assembly Government • research organisations that have done nationally recognised research in relation to the topics covered in the guideline. London: NICE. see Chapter 4). • Clinical guidelines – Type 1 diabetes: diagnosis and management of type 1 diabetes in children.

22. searches were not date specific. The most recent search conducted for the three Cochrane databases (Cochrane Central Register of Controlled Trials. theses and unpublished trials).1 summarises the key stages of the guideline development process and which version of the process was followed at each stage. Language restrictions were not applied to searches. etc. Additionally. Hand searching of journals not indexed on the databases was not undertaken. Towards the end of the guideline development process. although publications in languages other than English were not appraised. Searches to identify economic studies were undertaken using the above databases and the NHS Economic Evaluation Database (NHS EED). Table 1.7 Guideline methodology This guideline was developed in accordance with the NICE guideline development process outlined in the 2005 technical manual21 and the 2006 and 2007 editions of the NICE guidelines manual. Embase (1980 onwards). The reference lists in these guidelines were checked against subsequent searches to identify missing evidence. 1982 onwards). thereby including evidence published and included in the databases up to 21 March 2007. abstracts. Cumulative Index to Nursing and Allied Health Literature (CINAHL. There was no systematic attempt to search grey literature (conferences.1 Stages in the NICE guideline development process and the versions followed at each stage Stage 2005 version21 Scoping the guideline (determining what the guideline would and would not cover)  Preparing the work plan (agreeing timelines. and PsycINFO (1967 onwards). The clinical questions are presented in Appendix B. milestones. searches were updated and re-executed. stakeholder organisations were invited to submit evidence for consideration by the GDG provided it was relevant to the topics included in the scope and of equivalent or better quality than evidence identified by the search strategies. Cochrane Database of Systematic Reviews and the Database of Abstracts of Reviews of Effects) was undertaken in Quarter 1. Both generic and specially developed methodological search filters were used appropriately. guideline development  group constitution. 2007. Relevant published evidence to inform the guideline development process and answer the clinical questions was identified by systematic search strategies.23 Table 1. national and international) produced by other development groups.) Forming and running the guideline development group  Developing clinical questions  Identifying the evidence Reviewing and grading the evidence Incorporating health economics Making group decisions and reaching consensus Linking guidance to other NICE guidance Creating guideline recommendations Developing clinical audit criteria Writing the guideline Validation (stakeholder consultation on the draft guideline) Declaration of interestsa  a 2006 version22 2007 version23            The process for declaring interests was extended in November 2006 to cover NCC-WCH staff and to include personal family interests. Literature search strategy Initial scoping searches were executed to identify relevant guidelines (local. Unless advised by the GDG. 4 .Diabetes in pregnancy 1. Search strategies combined relevant controlled vocabulary and natural language in an effort to balance sensitivity and specificity. Systematic searches to answer the clinical questions formulated and agreed by the GDG were executed using the following databases via the ‘Ovid’ platform: Medline (1966 onwards).

studies rated as ‘−’ should not be used as a basis for making a recommendation. meta-analysis or RCT existed in relation to a question. Table 1. Usually. test evaluation studies examining the performance of the test were used if the effectiveness (accuracy) of the test was required. high-quality case–control or cohort studies with a very low risk of confounding. the highest available level of evidence was selected. The type of clinical question dictates the highest level of evidence that may be sought.2 Level 1++ 1+ 1− 2++ Levels of evidence for intervention studies Source of evidence High-quality meta-analyses. or RCTs with a very low risk of bias Well-conducted meta-analyses. for example. other appropriate experimental or observational studies were sought. bias or chance and a high probability that the relationship is causal Well-conducted case–control or cohort studies with a low risk of confounding. specificity = d/(b+d). PPV = a/(a+b). Studies of poor quality were rated as ‘−’.Introduction Evidence published after this date has not been included in the guideline. bias or chance and a significant risk that the relationship is not causal Non-analytical studies (for example. but they can be used to inform recommendations.3 Test positive ’2 × 2’ table for calculation of diagnostic accuracy parameters Reference standard positive a (true positive) c (false negative) a+c Reference standard negative Total b (false positive) d (true negative) b+d a+b c+d a+b+c+d = N (total number of tests in study) Test negative Total Sensitivity = a/(a+c). formal consensus 2+ 2− 3 4 Table 1. systematic reviews of RCTs. NPV = d/(c+d) 5 . For issues of therapy or treatment. bias or chance and a moderate probability that the relationship is causal Case–control or cohort studies with a high risk of confounding. sensitivity. but where an evaluation of the effectiveness of the test in the clinical management of patients and the outcome of disease was required. For each clinical question. In assessing the quality of the evidence. or RCTs with a low risk of bias Meta-analyses. systematic reviews of randomised controlled trials (RCTs). studies of a weaker design were not considered. case reports. each study was assigned a quality rating coded as ‘++’. if a systematic review. systematic reviews of RCTs. the highest possible evidence level (EL) is a well-conducted systematic review or meta-analysis of randomised controlled trials (RCTs. Where appropriate. This date should be considered the starting point for searching for new evidence for future updates to this guideline. case series) Expert opinion. For issues of prognosis.22 This system reflects the susceptibility to bias that is inherent in particular study designs. Appraisal and synthesis of clinical effectiveness evidence Evidence relating to clinical effectiveness was reviewed using established guides24–30 and classified using the established hierarchical system presented in Table 1. positive predictive values (PPVs) and negative predictive values (NPVs) were calculated or quoted where possible (see Table 1. evidence from RCTs or cohort studies was optimal. meta-analyses and RCTs did not exist. the highest possible level of evidence is a cohort study (EL = 2).3). EL = 1++) or an individual RCT (EL = 1+). For studies evaluating the accuracy of a diagnostic test.2. including the methodological filters employed. ‘+’ or ‘−’. For diagnostic tests. are provided on the accompanying CD-ROM. Further details of the search strategies. or RCTs with a high risk of bias High-quality systematic reviews of case–control or cohort studies. A level of evidence was assigned to each study appraised during the development of the guideline. Where systematic reviews. specificity.

which started in 2002. the effectiveness of interventions has been assessed against the following outcome domains. incontinence) – maternal diabetic complications (glycaemic control (glycosylated haemoglobin. admission to intensive care unit. macrovascular disease) – development of type 2 diabetes. i. • Maternal outcomes – preterm birth – mode of birth (spontaneous vaginal.22 Clinical evidence for individual studies was extracted into evidence tables (provided on the accompanying CD-ROM) and a brief description of each study was included in the guideline text. polycythaemia. infection. it is less appropriate for studies reporting accuracy of diagnostic tests.32 • a descriptive study of 3808 pregnancies followed until 28 days after birth in women with preexisting diabetes who booked or gave birth between 1 March 2002 and 28 February 20032 • a national confidential enquiry of demographic. social and lifestyle factors and clinical care in 442 pregnancies complicated by pre-existing diabetes. the guideline makes separate recommendations for women with pre-existing diabetes (type 1 diabetes. special care unit or transitional care unit – hypoglycaemia.e. low birthweight – shoulder dystocia. hypocalcaemia. thrombosis. wound healing – maternal death – maternal obstetric complications (haemorrhage. gestational diabetes was excluded). The term ‘women’ is used in the guideline to refer to all females of childbearing age. hypoxic ischaemic encephalopathy. stillbirth. nerve palsy) – admission to intensive care unit. type 2 diabetes and other forms of diabetes. 6 . which assessed the quality of maternity service provision against standards set out in the NSF for diabetes20. neonatal or infant death – congenital malformation – macrosomia. birth trauma (bone fracture. jaundice. resuscitation. retinopathy.4). hypoglycaemic episodes. focused on pre-existing diabetes (type 1 and type 2 diabetes.Diabetes in pregnancy The system described above covers studies of treatment effectiveness. In the absence of a validated ranking system for this type of test. Specific considerations for this guideline Where the evidence supports it. NICE has developed a hierarchy of evidence that takes into account the various factors likely to affect the validity of these studies (see Table 1. sepsis. instrumental.33 Results from the CEMACH diabetes in pregnancy programme were considered systematically by the GDG alongside other evidence and particularly as an indicator of current clinical practice. For this guideline. diabetic ketoacidosis (DKA). transient heart failure. respiratory distress. nephropathy. high-dependency unit. including young women who have not yet transferred from paediatric to adult services. Quantitative synthesis (metaanalysis) was not performed for this guideline because there were no clinical questions for which sufficient numbers of similar studies were identified to merit such analysis. small for gestational age (SGA). The body of evidence identified for each clinical question was synthesised qualitatively in clinical evidence statements that accurately reflected the evidence. • Neonatal outcomes – miscarriage. such as maturity-onset diabetes of the young)31 and gestational diabetes. The three parts of the programme were: • a survey of diabetes maternity services. However. CEMACH has conducted a three-part enquiry programme relating to diabetes in pregnancy. The programme. caesarean section) – mode of infant feeding – maternal health-related quality of life (validated questionnaire) – maternal satisfaction with experience of pregnancy and birth – perineal trauma. impairment of neurodevelopment. HbA1c).

bench research or ‘first principles’ b c Homogeneity means there are no or minor variations in the directions and degrees of results between individual studies that are included in the systematic review. systematic reviews of level-2 studies Level-3 studiesd. Level-3 studies are studies that have at least two or three of the features listed above. The results of each economic analysis are summarised briefly in the guideline text. d Health economics considerations The aims of the economic input to the guideline were to inform the GDG of potential economic issues relating to the management of diabetes and its complications from preconception to the postnatal period. Health economic considerations were aided by original economic analysis undertaken as part of the development of the guideline where robust clinical effectiveness data were available and UK cost data could be obtained.9) • treatment for gestational diabetes (see Section 4. The GDG prioritised a number of clinical questions where it was thought that economic considerations would be particularly important in formulating recommendations.4 Level Ia Ib II III IV a Levels of evidence for studies of the accuracy of diagnostic tests Type of evidence Systematic review (with homogeneity)a of level-1 studiesb Level-1 studiesb Level-2 studiesc. A systematic search for published economic evidence was undertaken for these questions.1). or where the ‘testing’ affects the ‘reference’) • the comparison between the test and reference standard is not blind • case–control studies. Detailed descriptions of the methods used for assessing the cost-effectiveness of self-management programmes are presented in Appendix C. or based on physiology. For this guideline the areas prioritised for economic analysis were: • self-management programmes for women with diabetes who are planning a pregnancy (see Section 3.3) – this was addressed through a unified analysis of screening.7) • criteria for admission to a neonatal intensive care unit (NICU) or special care unit (see Section 7. Level-1 studies are studies that use a blind comparison of the test with a validated reference standard (gold standard) in a sample of patients that reflects the population to whom the test would apply. diagnosis and treatment for gestational diabetes are presented in Appendix D and those for assessing the cost-effectiveness of screening for congenital cardiac malformations are described in Appendix E. Level-2 studies are studies that have only one of the following: • narrow population (the sample does not reflect the population to whom the test would apply) • use a poor reference standard (defined as that where the ‘test’ is included in the ‘reference’. no standard system of grading the quality of evidence exists and included papers were assessed using a quality assessment checklist based on good practice in economic evaluation. and to ensure that recommendations represented cost-effective use of healthcare resources. For economic evaluations. The methods used for assessing the cost-effectiveness of screening.34 Reviews of the very limited relevant published economic literature are presented alongside the clinical reviews or as part of appendices detailing original economic analyses (see below).6) • monitoring fetal growth and wellbeing (see Section 5. 7 .Introduction Table 1. expert committee reports or opinions and/or clinical experience without explicit critical appraisal. diagnosis and treatment for gestational diabetes involving joint work with the NICE antenatal care GDG • screening for congenital malformations (see Section 5. systematic reviews of level-3 studies Consensus.

The GDG also identified five key priorities for research (again using criteria set out in the NICE guidelines manual). which (in accordance with the criteria specified in the NICE guidelines manual23) were those recommendations expected to have the biggest impact on care and outcomes for pregnant women with diabetes and their babies in the NHS as a whole. The shortlisted recommendations were discussed at subsequent GDG meetings. 3. The GDG members’ votes were collated and a shortlist of priority recommendations was obtained using the same criteria that were used to shortlist recommendations for clinical care.6. Each GDG member submitted an electronic form indicating their top five research recommendations in order of priority. The GDG carefully considered and responded to all of the comments received from stakeholders during the consultation periods.1. 3. In areas where no substantial clinical research evidence was identified. informal consensus methods were used by the GDG to agree clinical and cost-effectiveness evidence statements.9 Schedule for updating the guideline Clinical guidelines commissioned by NICE are published with a review date 4 years from the date of publication. Each GDG member submitted an electronic form indicating their top five recommendations in order of priority. formal consensus methods were used to consider all the clinical care recommendations and research recommendations that had been drafted previously. contraindications and use in pregnancy and lactation of drugs used in diabetes management and retinal assessment (specifically insulins. the information on the therapeutic indications. The health economics justification in areas of the guideline where the use of NHS resources (interventions) was considered was based on GDG consensus in relation to the likely cost-effectiveness implications of the recommendations.6 and to the footnotes of recommendations in Sections 2. which were reviewed independently by a guideline review panel convened by NICE. the GDG considered other evidence-based guidelines and consensus statements or used their collective experience to identify good practice.10. and the final selection was made by retaining the recommendations that had received most votes.6. recommendations for clinical care were derived using. are published on the NICE website. and the final selection was made by retaining the recommendations that had received most votes and distilling the important issues contained in some long recommendations into more succinct recommendations. 4. Changes have been made to the recommendations in Sections 2.3. the evidence that supported them.2. 4. In this revised reprint. Towards the end of the guideline development process. The GDG members’ votes were collated and a shortlist of priority recommendations was obtained by including all recommendations that had been voted for by at least one GDG member.3 and 5. 5. The GDG also identified areas where evidence to answer their clinical questions was lacking and used this information to formulate recommendations for future research. again using a variant of the nominal group technique.2. The comments and responses. Statements summarising the GDG’s interpretation of the evidence and any extrapolation from the evidence used to form recommendations were also prepared. The updated guideline will be available within 2 years of the start of the review process.3 and 8. In the first instance. The key priorities were selected using a variant of the nominal group technique. 1. Footnotes have been deleted from recommendations in Sections 2. and tropicamide) has been corrected to follow the relevant summaries of product characteristics (SPCs) (July 2008).1).Diabetes in pregnancy GDG interpretation of the evidence and formulation of recommendations For each clinical question. 3. Reviewing may begin earlier than 4 years if significant evidence that affects guideline recommendations is identified sooner. The shortlisted recommendations were discussed at subsequent GDG meetings. 1. the oral hypoglycaemic agents metformin and glibenclamide. The GDG identified nine key priorities for implementation (key recommendations). 8 .8 Stakeholder involvement in the guideline development process Registered stakeholder organisations were invited to comment on the scope of the guideline during the scoping stage of development and on the evidence and recommendations in the validation stage (see Table 1.4.2 and 3. which were the most important research recommendations. and linked explicitly to.

women who are suspected of having diabetic ketoacidosis should be admitted immediately for level 2 critical care. 2. 9 . Management of diabetes during pregnancy Women with insulin-treated diabetes should be advised of the risks of hypoglycaemia and hypoglycaemia unawareness in pregnancy. women with diabetes should aim to keep fasting blood glucose between 3. It is important to explain that risks can be reduced but not eliminated.2 Summary of recommendations and algorithm Key priorities for implementation (key recommendations) Chapter 3 Preconception care Outcomes and risks for the woman and baby Women with diabetes who are planning to become pregnant should be informed that establishing good glycaemic control before conception and continuing this throughout pregnancy will reduce the risk of miscarriage.8 mmol/litre during pregnancy. including support for a single failing organ system or postoperative care and those ‘stepping down’ from higher levels of care. stillbirth and neonatal death. particularly in the first trimester. Importance of planning pregnancy and the role of contraception The importance of avoiding unplanned pregnancy should be an essential component of diabetes education from adolescence for women with diabetes.  Level 2 critical care is defined as care for patients requiring detailed observation or intervention. diet and exercise) and offered a fasting plasma glucose measurement (but not an oral glucose tolerance test) at the 6 week postnatal check and annually thereafter. where they can receive both medical and obstetric care. Self-management programmes Women with diabetes who are planning to become pregnant should be offered preconception care and advice before discontinuing contraception.9 mmol/litre and 1 hour postprandial blood glucose below 7. Chapter 8 Postnatal care Information and follow-up after birth Women who were diagnosed with gestational diabetes should be offered lifestyle advice (including weight control. Chapter 7 Neonatal care Initial assessment and criteria for admission to intensive/special care Babies of women with diabetes should be kept with their mothers unless there is a clinical complication or there are abnormal clinical signs that warrant admission for intensive or special care. During pregnancy.5 and 5.1 Chapter 5 Antenatal care Target ranges for blood glucose during pregnancy If it is safely achievable. Screening for congenital malformations Women with diabetes should be offered antenatal examination of the four chamber view of the fetal heart and outflow tracts at 18–20 weeks. congenital malformation.

body weight and exercise Women with diabetes who are planning to become pregnant should be offered individualised dietary advice. Women with diabetes who are planning to become pregnant and who have a body mass index above 27 kg/m² should be offered advice on how to lose weight in line with ‘Obesity: guidance on the prevention. Diet. medications for diabetes (including insulin regimens for insulin-treated diabetes) and medications for complications of diabetes will need to be reviewed before and during pregnancy • that additional time and effort is required to manage diabetes during pregnancy and that there will be frequent contact with healthcare professionals. 10 .  Diabetes Control and Complications Trial (DCCT)-aligned haemoglobin A1c (HbA1c) test.nice. Women with diabetes who are planning to become pregnant should be advised to take folic acid (5 mg/day) until 12 weeks of gestation to reduce the risk of having a baby with a neural tube defect. dietary supplements. The information should cover: • • • • • • • • • the role of diet. stillbirth and neonatal death. glucose monitoring. congenital malformation. advice and support that will help to reduce the risks of adverse pregnancy outcomes for mother and baby. Women with diabetes who are planning to become pregnant should be advised: • that the risks associated with pregnancies complicated by diabetes increase with the duration of diabetes • to use contraception until good glycaemic control (assessed by HbA1c) has been established • that glycaemic targets.Diabetes in pregnancy 2. induction of labour and caesarean section the need for assessment of diabetic retinopathy before and during pregnancy the need for assessment of diabetic nephropathy before pregnancy the importance of maternal glycaemic control during labour and birth and early feeding of the baby in order to reduce the risk of neonatal hypoglycaemia the possibility of transient morbidity in the baby during the neonatal period.uk/CG043. Women should be given information about the local arrangements for support. Women with diabetes who are planning to become pregnant should be informed that establishing good glycaemic control before conception and continuing this throughout pregnancy will reduce the risk of miscarriage. identification. body weight and exercise the risks of hypoglycaemia and hypoglycaemia unawareness during pregnancy how nausea and vomiting in pregnancy can affect glycaemic control the increased risk of having a baby who is large for gestational age.2 Summary of recommendations Chapter 3 Preconception care Outcomes and risks for the woman and baby Healthcare professionals should seek to empower women with diabetes to make the experience of pregnancy and childbirth a positive one by providing information.org. which may require admission to the neonatal unit the risk of the baby developing obesity and/or diabetes in later life. including emergency contact numbers. Women with diabetes who are planning to become pregnant and their families should be offered information about how diabetes affects pregnancy and how pregnancy affects diabetes. It is important to explain that risks can be reduced but not eliminated. available from www. The importance of planning pregnancy and the role of contraception The importance of avoiding unplanned pregnancy should be an essential component of diabetes education from adolescence for women with diabetes. which increases the likelihood of birth trauma. assessment and management of overweight and obesity in adults and children’ (NICE clinical guideline 43).

Women with insulin-treated diabetes who are planning to become pregnant should be informed that there is insufficient evidence about the use of long-acting insulin analogues during pregnancy. available from www. Healthcare professionals should be aware that data from clinical trials and other sources do not suggest that the rapid-acting insulin analogues (aspart and lispro) adversely affect the pregnancy or the health of the fetus or newborn baby. The safety of medications for diabetic complications before and during pregnancy Angiotensin-converting enzyme inhibitors and angiotensin-II receptor antagonists should be discontinued before conception or as soon as pregnancy is confirmed. Women with diabetes who are planning to become pregnant and who require intensification of hypoglycaemic therapy should be advised to increase the frequency of self-monitoring of blood glucose to include fasting and a mixture of pre.nice. Alternative antihypertensive agents suitable for use during pregnancy should be substituted. available from www. All other oral hypoglycaemic agents should be discontinued before pregnancy and insulin substituted. from adolescence. including their diabetes care team. taking into account the risk of hypoglycaemia.  Metformin is used in UK clinical practice in the management of diabetes in pregnancy and lactation.uk/TA060.and postprandial levels. 11 . updates the information on type 2 diabetes in this technology appraisal. If it is safely achievable. Women with diabetes who are planning to become pregnant should be offered a meter for selfmonitoring of blood glucose. The SPC advises that when a patient plans to become pregnant and during pregnancy.uk/CG066. Women with type 1 diabetes who are planning to become pregnant should be offered ketone testing strips and advised to test for ketonuria or ketonaemia if they become hyperglycaemic or unwell. Preconception care for women with diabetes should be given in a supportive environment and the woman’s partner or other family member should be encouraged to attend.org. Removing barriers to the uptake of preconception care and when to offer information Women with diabetes should be informed about the benefits of preconception glycaemic control at each contact with healthcare professionals. This evidence is not currently reflected in the SPC (July 2008). Self-management programmes Women with diabetes who are planning to become pregnant should be offered a structured education programme as soon as possible if they have not already attended one (see ‘Guidance on the use of patient-education models for diabetes’ [NICE technology appraisal guidance 60].nice. Women should be reassured that any reduction in HbA1c towards the target of 6.Summary of recommendations and algorithm Target ranges for blood glucose in the preconception period Individualised targets for self-monitoring of blood glucose should be agreed with women who have diabetes and are planning to become pregnant.  ‘Type 2 diabetes: the management of type 2 diabetes’ (NICE clinical guideline 66). The intentions of women with diabetes regarding pregnancy and contraceptive use should be documented at each contact with their diabetes care team from adolescence. Monitoring blood glucose and ketones in the preconception period Women with diabetes who are planning to become pregnant should be offered monthly measurement of HbA1c. Statins should be discontinued before pregnancy or as soon as pregnancy is confirmed.1% is likely to reduce the risk of congenital malformations. Informed consent on the use of metformin in these situations should be obtained and documented. There is strong evidence for its effectiveness and safety. women with diabetes who are planning to become pregnant should aim to maintain their HbA1c below 6. Women with diabetes whose HbA1c is above 10% should be strongly advised to avoid pregnancy. Therefore isophane insulin (also known as NPH insulin) remains the first choice for long-acting insulin during pregnancy. diabetes should not be treated with metformin but insulin should be used to maintain blood glucose levels.1%. when the likely benefits from improved glycaemic control outweigh the potential for harm.org. The safety of medications for diabetes before and during pregnancy Women with diabetes may be advised to use metformin as an adjunct or alternative to insulin in the preconception period and during pregnancy.

Pakistan or Bangladesh) – black Caribbean – Middle Eastern (specifically women whose country of family origin is Saudi Arabia. At the booking appointment. 12 . United Arab Emirates. Chapter 4 Gestational diabetes Risk factors for gestational diabetes Healthcare professionals should be aware that the following have been shown to be independent risk factors for gestational diabetes: • body mass index above 30 kg/m² • previous macrosomic baby weighing 4. Iraq. referral to a nephrologist should be considered before discontinuing contraception. Lebanon or Egypt). available from www. Kuwait. including a measure of microalbuminuria. Iraq. Qatar.5 kg or above • previous gestational diabetes • family history of diabetes (first-degree relative with diabetes) • family origin with a high prevalence of diabetes: – South Asian (specifically women whose country of family origin is India. Women with any one of these risk factors should be offered testing for gestational diabetes. Lebanon or Egypt). Screening. and treatment for gestational diabetes Screening for gestational diabetes using risk factors is recommended in a healthy population. Renal assessment in the preconception period Women with diabetes should be offered a renal assessment. Kuwait. Retinal assessment in the preconception period Women with diabetes seeking preconception care should be offered retinal assessment at their first appointment (unless an annual retinal assessment has occurred within the previous 6 months) and annually thereafter if no diabetic retinopathy is found.73 m². before discontinuing contraception. Syria. United Arab Emirates.org. Qatar. in line with the UK National Screening Committee’s recommendations for annual mydriatic twofield digital photographic screening as part of a systematic screening programme. Oman. the following risk factors for gestational diabetes should be determined: • • • • • body mass index above 30 kg/m² previous macrosomic baby weighing 4.  This recommendation is taken from ‘Antenatal care: routine care for the healthy pregnant woman’ (NICE clinical guideline 62).Diabetes in pregnancy Women with diabetes who are planning to become pregnant should be offered preconception care and advice before discontinuing contraception.uk/CG062. or the estimated glomerular filtration rate (eGFR) is less than 45 ml/minute/1. Oman.nice. Jordan. Jordan. Women with diabetes who are planning to become pregnant should be advised to defer rapid optimisation of glycaemic control until after retinal assessment and treatment have been completed. Retinal assessment should be carried out by digital imaging with mydriasis using tropicamide.5 kg or above previous gestational diabetes family history of diabetes (first-degree relative with diabetes) family origin with a high prevalence of diabetes: – South Asian (specifically women whose country of family origin is India. If serum creatinine is abnormal (120 micromol/litre or more). Syria. diagnosis. Pakistan or Bangladesh) – black Caribbean – Middle Eastern (specifically women whose country of family origin is Saudi Arabia.

random blood glucose. neonatal hypoglycaemia and perinatal death. Informed consent on the use of glibenclamide in pregnancy should be obtained and documented. Women with gestational diabetes should be advised to choose. carbohydrates from low glycaemic index sources. World Health Organization Department of Noncommunicable Disease Surveillance (1999) Definition.  Fasting plasma venous glucose concentration greater than or equal to 7. available from www. 13 .org.nice. Women with gestational diabetes should be offered information covering: • the role of diet. gestational diabetes will respond to changes in diet and exercise • some women (between 10% and 20%) will need oral hypoglycaemic agents or insulin therapy if diet and exercise are not effective in controlling gestational diabetes • if gestational diabetes is not detected and controlled there is a small risk of birth complications such as shoulder dystocia • a diagnosis of gestational diabetes may lead to increased monitoring and interventions during both pregnancy and labour. trauma during birth (to themselves and the baby). Hypoglycaemic therapy should be considered for women with gestational diabetes if diet and exercise fail to maintain blood glucose targets during a period of 1–2 weeks. Targets for blood glucose control should be determined in the same way as for women with preexisting diabetes. the individual woman. lean proteins including oily fish and a balance of polyunsaturated fats and monounsaturated fats. Part 1: diagnosis and classification of diabetes mellitus. Hypoglycaemic therapy for women with gestational diabetes (which may include regular insulin. Women with gestational diabetes should be instructed in self-monitoring of blood glucose. Screening for gestational diabetes using fasting plasma glucose.  Glibenclamide is used in UK clinical practice in the management of diabetes in pregnancy and lactation. Informed consent on the use of metformin in these situations should be obtained and documented. diagnosis and classification of diabetes mellitus and its complications. The 2 hour 75 g oral glucose tolerance test (OGTT) should be used to test for gestational diabetes and diagnosis made using the criteria defined by the World Health Organization. Women with gestational diabetes whose pre-pregnancy body mass index was above 27 kg/m² should be advised to restrict calorie intake (to 25 kcal/kg/day or less) and to take moderate exercise (of at least 30 minutes daily). induction of labour or caesarean section. where possible. Geneva: World Health Organization.8 mmol/litre. Hypoglycaemic therapy should be considered for women with gestational diabetes if ultrasound investigation suggests incipient fetal macrosomia (abdominal circumference above the 70th percentile) at diagnosis. women should be informed that: • in most women. glucose challenge test and urinalysis for glucose should not be undertaken. diabetes should not be treated with metformin but insulin should be used to maintain blood glucose levels. There is strong evidence for its effectiveness and safety. and a further OGTT at 28 weeks if the results are normal. There is strong evidence for its effectiveness and safety. induction of labour and caesarean section • the importance of maternal glycaemic control during labour and birth and early feeding of the baby in order to reduce the risk of neonatal hypoglycaemia • the possibility of transient morbidity in the baby during the neonatal period. This evidence is not currently reflected in the SPC (July 2008).Summary of recommendations and algorithm In order to make an informed decision about screening and testing for gestational diabetes.uk/CG062. and acceptability to.  This recommendation is taken from ‘Antenatal care: routine care for the healthy pregnant woman’ (NICE clinical guideline 62). Women with any of the other risk factors for gestational diabetes should be offered an OGTT at 24–28 weeks. Report of a WHO consultation. Women with gestational diabetes should be informed that good glycaemic control throughout pregnancy will reduce the risk of fetal macrosomia. The SPC advises that when a patient plans to become pregnant and during pregnancy. The SPC advises that glibenclamide is contraindicated in pregnancy. This evidence is not currently reflected in the SPC (July 2008). rapid-acting insulin analogues [aspart and lispro] and/or hypoglycaemic agents [metformin and glibenclamide] should be tailored to the glycaemic profile of. Women who have had gestational diabetes in a previous pregnancy should be offered early self-monitoring of blood glucose or OGTT at 16–18 weeks. body weight and exercise • the increased risk of having a baby who is large for gestational age.  Metformin is used in UK clinical practice in the management of diabetes in pregnancy and lactation. which increases the likelihood of birth trauma. which may require admission to the neonatal unit • the risk of the baby developing obesity and/or diabetes in later life.0 mmol/lite or 2 hour plasma venous glucose concentration greater than or equal to 7.

During pregnancy. women who are suspected of having diabetic ketoacidosis should be admitted immediately for level 2 critical care. where they can receive both medical and obstetric care. 14 . particularly in the first trimester. During pregnancy. During pregnancy. women and their partners or other family members should be instructed in their use. an additional retinal assessment should be performed at 16–20 weeks. women with type 1 diabetes who become unwell should have diabetic ketoacidosis excluded as a matter of urgency. If retinal assessment has not been performed in the preceding 12 months. Women with type 1 diabetes who are pregnant should be offered ketone testing strips and advised to test for ketonuria or ketonaemia if they become hyperglycaemic or unwell. Management of diabetes during pregnancy Healthcare professionals should be aware that the rapid-acting insulin analogues (aspart and lispro) have advantages over soluble human insulin during pregnancy and should consider their use. Monitoring blood glucose and ketones during pregnancy Women with diabetes should be advised to test fasting blood glucose levels and blood glucose levels 1 hour after every meal during pregnancy. HbA1c should not be used routinely for assessing glycaemic control in the second and third trimesters of pregnancy. it should be offered as soon as possible after the first contact in pregnancy in women with pre-existing diabetes. Diabetic retinopathy should not be considered a contraindication to rapid optimisation of glycaemic control in women who present with a high HbA1c in early pregnancy. Women with insulin-treated diabetes should be advised to test blood glucose levels before going to bed at night during pregnancy. women with diabetes should aim to keep fasting blood glucose between 3. Retinal assessment during pregnancy Pregnant women with pre-existing diabetes should be offered retinal assessment by digital imaging with mydriasis using tropicamide following their first antenatal clinic appointment and again at 28 weeks if the first assessment is normal.9 mmol/litre and 1 hour postprandial blood glucose below 7. including support for a single failing organ system or postoperative care and those ‘stepping down’ from higher levels of care. If it is safely achievable. women with insulin-treated diabetes should be provided with a concentrated glucose solution and women with type 1 diabetes should also be given glucagon. Women who have preproliferative diabetic retinopathy diagnosed during pregnancy should have ophthalmological follow-up for at least 6 months following the birth of the baby. Women with insulin-treated diabetes should be advised of the risks of hypoglycaemia and hypoglycaemia unawareness in pregnancy. During pregnancy. ‘disabling hypoglycaemia’ means the repeated and unpredicted occurrence of hypoglycaemia requiring third-party assistance that results in continuing anxiety about recurrence and is associated with significant adverse effect on quality of life  Level 2 critical care is defined as care for patients requiring detailed observation or intervention.Diabetes in pregnancy Chapter 5 Antenatal care Target ranges for blood glucose during pregnancy Individualised targets for self-monitoring of blood glucose should be agreed with women with diabetes in pregnancy. taking into account the risk of hypoglycaemia. Diabetic retinopathy should not be considered a contraindication to vaginal birth.5 and 5.  For the purpose of this guidance.8 mmol/litre during pregnancy. If any diabetic retinopathy is present. women with insulin-treated diabetes should be offered continuous subcutaneous insulin infusion (CSII or insulin pump therapy) if adequate glycaemic control is not obtained by multiple daily injections of insulin without significant disabling hypoglycaemia.

unless there is a risk of intrauterine growth restriction. Antenatal appointments for women with diabetes should provide care specifically for women with diabetes. referral to a nephrologist should be considered (eGFR should not be used during pregnancy). Preterm labour in women with diabetes Diabetes should not be considered a contraindication to antenatal steroids for fetal lung maturation or to tocolysis. or by elective caesarean section if indicated. Routine monitoring of fetal wellbeing before 38 weeks is not recommended in pregnant women with diabetes. Betamimetic drugs should not be used for tocolysis in women with diabetes.nice. If general anaesthesia is used for the birth in women with diabetes. At each appointment women should be offered ongoing opportunities for information and education. blood glucose should be monitored regularly (every 30 minutes) from induction of general anaesthesia until after the baby is born and the woman is fully conscious. 15 .7 describes where care for women with diabetes differs from routine antenatal care. Diabetes should not in itself be considered a contraindication to attempting vaginal birth after a previous caesarean section. after 38 completed weeks. Women with diabetes should have contact with the diabetes care team for assessment of glycaemic control every 1–2 weeks throughout pregnancy. induction of labour and caesarean section. Analgesia and anaesthesia Women with diabetes and comorbidities such as obesity or autonomic neuropathy should be offered an anaesthetic assessment in the third trimester of pregnancy. If serum creatinine is abnormal (120 micromol/litre or more) or if total protein excretion exceeds 2 g/day. Screening for congenital malformations Women with diabetes should be offered antenatal examination of the four chamber view of the fetal heart and outflow tracts at 18–20 weeks.uk/CG062). in addition to the care provided routinely for healthy pregnant women (see ‘Antenatal care: routine care for the healthy pregnant woman’ [NICE clinical guideline 62].org. it should be arranged at the first contact in pregnancy. available from www. Table 5. Monitoring fetal growth and wellbeing Pregnant women with diabetes should be offered ultrasound monitoring of fetal growth and amniotic fluid volume every 4 weeks from 28 to 36 weeks.Summary of recommendations and algorithm Renal assessment during pregnancy If renal assessment has not been undertaken in the preceding 12 months in women with pre-existing diabetes. Timetable of antenatal appointments Women with diabetes who are pregnant should be offered immediate contact with a joint diabetes and antenatal clinic. Pregnant women with diabetes who have an ultrasound-diagnosed macrosomic fetus should be informed of the risks and benefits of vaginal birth. Women with diabetes and a risk of intrauterine growth restriction (macrovascular disease and/or nephropathy) will require an individualised approach to monitoring fetal growth and wellbeing. Thromboprophylaxis should be considered for women with proteinuria above 5 g/day (macroalbuminuria). Chapter 6 Intrapartum care Timing and mode of birth Pregnant women with diabetes who have a normally grown fetus should be offered elective birth through induction of labour. Women with insulin-treated diabetes who are receiving steroids for fetal lung maturation should have additional insulin according to an agreed protocol and should be closely monitored.

education and advice about how diabetes will affect the (ideally by 10 weeks) pregnancy. including appointments at 25 weeks (for nulliparous women) and 34 weeks. Take a clinical history to establish the extent of diabetes-related complications. and start regular tests of fetal wellbeing for women with diabetes who are awaiting spontaneous labour. Offer tests of fetal wellbeing. mode and management of birth • analgesia and anaesthesia • changes to hypoglycaemic therapy during and after birth • management of the baby after birth • initiation of breastfeeding and the effect of breastfeeding on glycaemic control • contraception and follow-up. Confirm viability of pregnancy and gestational age. Glycaemic control during labour and birth During labour and birth. 32 weeks 36 weeks 38 weeks 39 weeks 40 weeks 41 weeks a Women with diabetes should also receive routine care according to the schedule of appointments in ‘Antenatal care: routine care for the healthy pregnant woman’ (NICE clinical guideline 62). Intravenous dextrose and insulin infusion is recommended during labour and birth for women with diabetes whose blood glucose is not maintained at between 4 and 7 mmol/litre. Offer four chamber view of the fetal heart and outflow tracts plus scans that would be offered at 18–20 weeks as part of routine antenatal care. Women with type 1 diabetes should be considered for intravenous dextrose and insulin infusion from the onset of established labour. 16 . Offer retinal and/or renal assessment if these have not been undertaken in the previous 12 months. 7–9 weeks Booking appointment Discuss information.Diabetes in pregnancy Table 5. Chapter 7 Neonatal care Initial assessment and criteria for admission to intensive or special care Women with diabetes should be advised to give birth in hospitals where advanced neonatal resuscitation skills are available 24 hours a day. birth and early parenting (such as breastfeeding and initial care of the baby). Review medications for diabetes and its complications. Offer induction of labour. Offer ultrasound monitoring of fetal growth and amniotic fluid volume. Offer retinal assessment to women with pre-existing diabetes who showed no diabetic retinopathy at their first antenatal clinic visit. but with the exception of the appointment for nulliparous women at 31 weeks. or caesarean section if indicated. Offer tests of fetal wellbeing. 16 weeks 20 weeks 28 weeks Offer retinal assessment at 16–20 weeks to women with pre-existing diabetes who showed signs of diabetic retinopathy at the first antenatal appointment. advice and support in relation to optimising glycaemic control. Offer ultrasound monitoring of fetal growth and amniotic fluid volume Offer information and advice about: • timing. Offer ultrasound monitoring of fetal growth and amniotic fluid volume.7 Specific antenatal care for women with diabetes Appointment First appointment (joint diabetes and antenatal clinic) Care for women with diabetes during pregnancya Offer information. Offer tests of fetal wellbeing. Offer to nulliparous women all investigations that would be offered at 31 weeks as part of routine antenatal care. Babies of women with diabetes should be kept with their mothers unless there is a clinical complication or there are abnormal clinical signs that warrant admission for intensive or special care. capillary blood glucose should be monitored on an hourly basis in women with diabetes and maintained at between 4 and 7 mmol/litre.

Women who have been diagnosed with gestational diabetes should discontinue hypoglycaemic treatment immediately after birth. 17 . Informed consent on the use of metformin during lactation should be obtained and documented. There is strong evidence for its effectiveness and safety. Women with insulin-treated pre-existing diabetes should be informed that they are at increased risk of hypoglycaemia in the postnatal period.  Glibenclamide is used in UK clinical practice in the management of diabetes in pregnancy and lactation. Chapter 8 Postnatal care Breastfeeding and effects on glycaemic control Women with insulin-treated pre-existing diabetes should reduce their insulin immediately after birth and monitor their blood glucose levels carefully to establish the appropriate dose. This evidence is not currently reflected in the SPC (July 2008). and they should be advised to have a meal or snack available before or during feeds. detection and management of hypoglycaemia in babies of women with diabetes. additional measures such as tube feeding or intravenous dextrose should be given. Babies of women with diabetes should feed as soon as possible after birth (within 30 minutes) and then at frequent intervals (every 2–3 hours) until feeding maintains pre-feed blood glucose levels at a minimum of 2. Babies of women with diabetes should be admitted to the neonatal unit if they have: • • • • • • • • • hypoglycaemia associated with abnormal clinical signs respiratory distress signs of cardiac decompensation due to congenital heart disease or cardiomyopathy signs of neonatal encephalopathy signs of polycythaemia and are likely to need partial exchange transfusion need for intravenous fluids need for tube feeding (unless adequate support is available on the postnatal ward) jaundice requiring intense phototherapy and frequent monitoring of bilirubinaemia been born before 34 weeks (or between 34 and 36 weeks if dictated clinically by the initial assessment of the baby and feeding on the labour ward). Babies of women with diabetes should have an echocardiogram performed if they show clinical signs associated with congenital heart disease or cardiomyopathy. hyperbilirubinaemia. The SPC advises that metformin is contraindicated in lactation. if there are abnormal clinical signs or if the baby will not feed orally effectively. Informed consent on the use of glibenclamide during lactation should be obtained and documented. Babies of women with diabetes who present with clinical signs of hypoglycaemia should have their blood glucose tested and be treated with intravenous dextrose as soon as possible. There is strong evidence for its effectiveness and safety.Summary of recommendations and algorithm Blood glucose testing should be carried out routinely in babies of women with diabetes at 2–4 hours after birth. If blood glucose values are below 2. Additional measures should only be implemented if one or more of these criteria are met. Babies of women with diabetes should have their blood glucose tested using a quality-assured method validated for neonatal use (ward-based glucose electrode or laboratory analysis). Women with pre-existing type 2 diabetes who are breastfeeding can resume or continue to take metformin and glibenclamide immediately following birth but other oral hypoglycaemic agents should be avoided while breastfeeding. Blood tests for polycythaemia.  Metformin is used in UK clinical practice in the management of diabetes in pregnancy and lactation. This evidence is not currently reflected in the SPC (July 2008). hypocalcaemia and hypomagnesaemia should be carried out for babies with clinical signs. The SPC advises that there is insufficient/limited information on the excretion of glibenclamide in human or animal breast milk. including heart murmur. especially when breastfeeding.0 mmol/litre on two consecutive readings despite maximal support for feeding. Prevention and assessment of neonatal hypoglycaemia All maternity units should have a written policy for the prevention.0 mmol/litre. and not before healthcare professionals are satisfied that the babies are maintaining blood glucose levels and are feeding well. Babies of women with diabetes should not be transferred to community care until they are at least 24 hours old. The timing of the examination will depend on the clinical circumstances.

3 Key priorities for research Chapter 4 Gestational diabetes Screening. 18 . The new technology may identify women in whom short-term postprandial peaks of glycaemia are not detected by intermittent blood glucose testing. The aim of monitoring is to adjust insulin regimens to reduce the incidence of adverse outcomes of pregnancy (for example. A subsequent OGTT should be offered if the test results in early pregnancy are normal. Women with diabetes should be reminded of the importance of contraception and the need for preconception care when planning future pregnancies.Diabetes in pregnancy Women with diabetes who are breastfeeding should continue to avoid any drugs for the treatment of diabetes complications that were discontinued for safety reasons in the preconception period. Women who were diagnosed with gestational diabetes should have their blood glucose tested to exclude persisting hyperglycaemia before they are transferred to community care. Women who were diagnosed with gestational diabetes (including those with ongoing impaired glucose regulation) should be offered early self-monitoring of blood glucose or an OGTT in future pregnancies. and universal OGTT (with or without fasting)? Why this is important Following the Australian carbohydrate intolerance study in pregnant women (ACHOIS) it seems that systematic screening for gestational diabetes may be beneficial to the UK population. which have not been systematically evaluated for clinical and cost-effectiveness (including acceptability) within the UK. diagnosis and treatment for gestational diabetes What is the clinical and cost-effectiveness of the three main available screening techniques for gestational diabetes: risk factors. Women who were diagnosed with gestational diabetes should be offered lifestyle advice (including weight control. Research is needed to determine whether the technology is likely to have a place in the clinical management of diabetes in pregnancy. A multicentre randomised controlled trial is required to test the existing screening techniques. so there is currently no evidence to assess its effectiveness outside the laboratory situation. Chapter 5 Antenatal care Monitoring blood glucose and ketones during pregnancy How effective is ambulatory continuous blood glucose monitoring in pregnancies complicated by diabetes? Why this is important The technology for performing ambulatory continuous blood glucose monitoring is only just becoming available. Women who were diagnosed with gestational diabetes should be reminded of the symptoms of hyperglycaemia. Women who were diagnosed with gestational diabetes (including those with ongoing impaired glucose regulation) should be informed about the risks of gestational diabetes in future pregnancies and they should be offered screening (OGTT or fasting plasma glucose) for diabetes when planning future pregnancies. two-stage screening by the glucose challenge test and OGTT. 2. diet and exercise) and offered a fasting plasma glucose measurement (but not an OGTT) at the 6 week postnatal check and annually thereafter. fetal macrosomia. caesarean section and neonatal hypoglycaemia). so these outcomes should be assessed as part of the research. Information and follow-up after birth Women with pre-existing diabetes should be referred back to their routine diabetes care arrangements.

particularly with the availability of insulin analogues (which may have improved the effectiveness of intermittent insulin injections). what topics are covered).Summary of recommendations and algorithm Management of diabetes during pregnancy Do new-generation continuous subcutaneous insulin infusion (CSII) pumps offer an advantage over traditional intermittent insulin injections in terms of pregnancy outcomes in women with type 1 diabetes? Why this is important Randomised controlled trials have shown no advantage or disadvantage of using CSII pumps over intermittent insulin injections in pregnancy. whether 19 . Specifically it should evaluate different forms of content (i. neonatal hypoglycaemia and respiratory distress). Chapter 6 Intrapartum care Glycaemic control during labour and birth What is the optimal method for controlling glycaemia during labour and birth? Why this is important Epidemiological studies have shown that poor glycaemic control during labour and birth is associated with adverse neonatal outcomes (in particular.e. 2. Conventional tests of fetal wellbeing (umbilical artery Doppler ultrasound. Future research should seek to establish the clinical and cost-effectiveness of different models of preconception care and advice for women with preexisting diabetes. frequency and timing of contact with healthcare professionals (for example. Alternative approaches that include measurements of liquor erythropoietin and magnetic resonance imaging spectroscopy may be effective. The potential benefits of intermittent insulin injections and/or CSII over intravenous dextrose plus insulin during the intrapartum period include patient preference due to the psychological effect of the woman feeling in control of her diabetes and having increased mobility.4 Summary of research recommendations Chapter 3 Preconception care Self-management programmes What is the most clinically and cost-effective form of preconception care and advice for women with diabetes? Why this is important Preconception care and advice for women with pre-existing diabetes is recommended because a health economic analysis has demonstrated cost-effectiveness of attendance at a preconception clinic. but there is currently insufficient clinical evidence to evaluate them. A new generation of CSII pumps may offer technological advantages that would make a randomised controlled trial appropriate. Well-designed randomised controlled trials that are sufficiently powered are needed to determine whether these approaches are clinically and cost-effective. Monitoring fetal growth and wellbeing How can the fetus at risk of intrauterine death be identified in women with diabetes? Why this is important Unheralded intrauterine death remains a significant contributor to perinatal mortality in pregnancies complicated by diabetes. Randomised controlled trials are therefore needed to evaluate the safety of intermittent insulin injections and/or CSII during labour and birth compared with that of intravenous dextrose plus insulin. cardiotocography and other biophysical tests) have been shown to have poor sensitivity for predicting such events. no randomised controlled trials have compared the effectiveness of intermittent subcutaneous insulin injections and/or CSII with that of intravenous dextrose plus insulin during labour and birth. However. Due to limitations in the clinical evidence available to inform the health economic modelling it was not possible to establish the optimal form of preconception care and advice for this group of women.

Diabetes in pregnancy one long session is more clinically and cost-effective than a series of shorter sessions). and format (for example. The research should also seek to establish whether women with type 1 and type 2 diabetes have different needs in terms of preconception care and advice. Retinal assessment during pregnancy Should retinal assessment during pregnancy be offered to women diagnosed with gestational diabetes who are suspected of having pre-existing diabetes? Why this is important Women with gestational diabetes may have previously unrecognised type 2 diabetes with retinopathy. Chapter 5 Antenatal care Monitoring blood glucose and ketones during pregnancy How effective is ambulatory continuous blood glucose monitoring in pregnancies complicated by diabetes? Why this is important The technology for performing ambulatory continuous blood glucose monitoring is only just becoming available. The aim of monitoring is to adjust insulin regimens to reduce the incidence of adverse outcomes of pregnancy (for example. Chapter 4 Gestational diabetes Screening. diagnosis and treatment for gestational diabetes What is the clinical and cost-effectiveness of the three main available screening techniques for gestational diabetes: risk factors. two-stage screening by the glucose challenge test and OGTT. A new generation of CSII pumps may offer technological advantages that would make a randomised controlled trial appropriate. At present this is not screened for because of the difficulty in identifying these 20 . Research is needed to determine whether the technology is likely to have a place in the clinical management of diabetes in pregnancy. whether preconception care and advice provided by a multidisciplinary team is more clinically and cost-effective than contact with one healthcare professional). The new technology may identify women in whom short-term postprandial peaks of glycaemia are not detected by intermittent blood glucose testing. so these outcomes should be assessed as part of the research. caesarean section and neonatal hypoglycaemia). whether group sessions are more clinically and cost-effective than providing care and advice for each woman separately). fetal macrosomia. and universal OGTT (with or without fasting)? Why this is important Following the Australian carbohydrate intolerance study in pregnant women (ACHOIS) it seems that systematic screening for gestational diabetes may be beneficial to the UK population. which have not been systematically evaluated for clinical and cost-effectiveness (including acceptability) within the UK. which healthcare professionals should be involved (for example. so there is currently no evidence to assess its effectiveness outside the laboratory situation. particularly with the availability of insulin analogues (which may have improved the effectiveness of intermittent insulin injections). A multicentre randomised controlled trial is required to test the existing screening techniques. Management of diabetes during pregnancy Do new-generation continuous subcutaneous insulin infusion (CSII) pumps offer an advantage over traditional intermittent insulin injections in terms of pregnancy outcomes in women with type 1 diabetes? Why this is important Randomised controlled trials have shown no advantage or disadvantage of using CSII pumps over intermittent insulin injections in pregnancy. and how different models of care and advice compare to structured education programmes already offered to women with type 1 and type 2 diabetes.

How effective is transvaginal ultrasound for the detection of congenital malformations in women with diabetes and coexisting obesity? Why this is important Obstetric ultrasound signals are attenuated by the woman’s abdominal wall fat. Further research is. The benefit of recognising such women is that treatment for diabetic retinopathy is available and could prevent short. and this may limit the sensitivity of abdominal ultrasound screening for congenital malformations. This is important because women with diabetes are at increased risk of having a baby with a congenital malformation. needed to evaluate the diagnostic accuracy and affect on pregnancy outcomes of vaginal ultrasound in women with diabetes and coexisting obesity.Summary of recommendations and algorithm women amongst the larger group who have reversible gestational diabetes. However. 21 . The benefit of clinically and cost-effective prophylactic treatment would be to significantly improve pregnancy outcomes in this group of women. Vaginal ultrasound does not have this difficulty. Monitoring fetal growth and wellbeing How can the fetus at risk of intrauterine death be identified in women with diabetes? Why this is important Unheralded intrauterine death remains a significant contributor to perinatal mortality in pregnancies complicated by diabetes. two clinical studies have reported conflicting results in terms of whether levels of PAPP-A in women with type 1 diabetes are lower than those in other women. but there is currently no evidence that it is more effective than abdominal ultrasound. needed to evaluate the diagnostic accuracy and effect on pregnancy outcomes of screening tests for Down’s syndrome incorporating measurements of PAPP-A in women with pre-existing diabetes. cardiotocography and other biophysical tests) have been shown to have poor sensitivity for predicting such events. Conventional tests of fetal wellbeing (umbilical artery Doppler ultrasound. therefore. but it is not performed routinely in antenatal clinics for women with pre-existing diabetes because a place for prophylactic treatment of pre-eclampsia in microalbuminuria-positive women has not been investigated. Well-designed randomised controlled trials that are sufficiently powered are needed to determine whether these approaches are clinically and cost-effective. Research studies are. Alternative approaches that include measurements of liquor erythropoietin and magnetic resonance imaging spectroscopy may be effective. Current practice is to adjust PAPP-A measurements in women with diabetes on the assumption that their PAPP-A levels are indeed lower than those of other women. The research needed would be an observational study of retinal assessment in women newly diagnosed with gestational diabetes to determine whether there is a significant amount of retinal disease present. Many women with diabetes (and particularly women with type 2 diabetes) are obese. but there is currently insufficient clinical evidence to evaluate them. The severity of any abnormality detected might identify women most at risk for appropriate retinal assessment. therefore. Renal assessment during pregnancy Does identification of microalbuminuria during pregnancy offer the opportunity for appropriate pharmacological treatment to prevent progression to pre-eclampsia in women with pre-existing diabetes? Why this is important Microalbuminuria testing is available. Screening for congenital malformations How reliable is first-trimester screening for Down’s syndrome incorporating levels of pregnancyassociated plasma protein (PAPP-A) in women with pre-existing diabetes? Why this is important Several screening tests for Down’s syndrome incorporate measurements of PAPP-A.or long-term deterioration of visual acuity.

There is also a putative risk of precipitating uterine contractions through antenatal expression of colostrum and an accompanying release of oxytocin. Antenatal expression and storage of colostrum may. Encouraging women with diabetes to express and store colostrum before birth might be viewed as an additional barrier to breastfeeding in this group of women who already have lower breastfeeding rates than the general maternity population. Chapter 8 Postnatal care Information and follow-up after birth Are there suitable long-term pharmacological interventions to be recommended postnatally for women who have been diagnosed with gestational diabetes to prevent the onset of type 2 diabetes? Why this is important Oral hypoglycaemic agents such rosiglitazone and metformin offer the possibility of pharmacological treatment for prevention of progression to type 2 diabetes in women who have been diagnosed with gestational diabetes. therefore. and the cardiovascular effects of spinal anaesthesia and vasopressors on diabetic control.Diabetes in pregnancy Chapter 6 Intrapartum care Analgesia and anaesthesia What are the risks and benefits associated with analgesia and anaesthesia in women with diabetes? Why this is important The increasing number of women with diabetes and the high rate of intervention during birth emphasise the need for clinical studies to determine the most effective methods for analgesia and anaesthesia in this group of women. As yet there have been no clinical studies to 22 . no randomised controlled trials have compared the effectiveness of intermittent subcutaneous insulin injections and/or CSII with that of intravenous dextrose plus insulin during labour and birth. neonatal hypoglycaemia and respiratory distress). Randomised controlled trials are therefore needed to evaluate the safety of intermittent insulin injections and/or CSII during labour and birth compared with that of intravenous dextrose plus insulin. be of benefit to babies of women with diabetes. Glycaemic control during labour and birth What is the optimal method for controlling glycaemia during labour and birth? Why this is important Epidemiological studies have shown that poor glycaemic control during labour and birth is associated with adverse neonatal outcomes (in particular. The potential benefits of intermittent insulin injections and/or CSII over intravenous dextrose plus insulin during the intrapartum period include patient preference due to the psychological effect of the woman feeling in control of her diabetes and having increased mobility. However. These factors should be explored in the randomised controlled trials. The research studies should investigate the effect of analgesia during labour. Chapter 7 Neonatal care Prevention and assessment of neonatal hypoglycaemia Is systematic banking of colostrum antenatally of any benefit in pregnancies complicated by diabetes? Why this is important Babies of women with diabetes are at increased risk of neonatal hypoglycaemia and may need frequent early feeding to establish and maintain normoglycaemia. Randomised controlled trials are needed to determine whether this practice is clinically and cost-effective. Additionally. the opportunity for early skin-to-skin contact and initiation of breastfeeding is not always achieved in pregnancies complicated by diabetes because of the increased risk of neonatal complications requiring admission to intensive/special care. There have been no clinical studies to evaluate the effectiveness of antenatal banking of colostrum in women with diabetes.

23 . 2.Summary of recommendations and algorithm investigate the effectiveness of oral hypoglycaemic agents in this context.5 Algorithm The algorithm for specific antenatal care for women with diabetes on the following four pages is reproduced from the Quick Reference Guide version of this guideline (revised July 2008). Randomised controlled trials are needed to determine the clinical and cost-effectiveness of such treatments compared with diet and exercise.

Diabetes in pregnancy Diabetes in pregnancy Antenatal care Key: Appointment including specific diabetes care Appointment including routine care only No routine care appointment Antenatal care See also the NICE clinical guideline on antenatal care (www. Offer: ● immediate referral to a joint diabetes and antenatal clinic ● ● contact with the diabetes care team every 1–2 weeks to assess glycaemic control advice on where to have the birth. birth and early parenting (such as breastfeeding and initial care of the baby). which should be in a hospital with advanced neonatal resuscitation skills available 24 hours a day information and education at each appointment care specifically for women with diabetes. ● Take a clinical history. 16 weeks ● Offer retinal assessment to women with pre-existing diabetes who had signs of diabetic retinopathy at the first antenatal appointment (see box 10). education and advice about how diabetes will affect pregnancy. advice and support on glycaemic control (see boxes 7–9). ● Offer retinal and renal assessment if these have not been performed in the previous 12 months (see boxes 10–11). Booking appointment (ideally by 10 weeks) ● Discuss information. 20 weeks ● Offer four-chamber view of the fetal heart and outflow tracts (see box 12).org. 25 weeks ● Offer routine care only (appointment for nulliparous women). ● Review medications. ● ● Specific antenatal care for women with diabetes First appointment (joint diabetes and antenatal clinic) ● Offer information. as described below. continued on page 11 10 NICE clinical guideline 63 Quick reference guide 24 . ● Offer scans that would be offered at 18–20 weeks in routine antenatal care.nice. 7–9 weeks ● Confirm viability of pregnancy and gestational age. in addition to routine antenatal care.uk/CG062).

39 weeks ● Offer tests of fetal well-being for women waiting for spontaneous labour. 34 weeks ● Offer routine care only. 41 weeks ● Offer tests of fetal well-being for women waiting for spontaneous labour. 38 weeks ● Offer induction of labour. 32 weeks ● Offer ultrasound monitoring of fetal growth and amniotic fluid volume. or caesarean section if indicated.Summary of recommendations and algorithm Diabetes in pregnancy Antenatal care 28 weeks ● Offer ultrasound monitoring of fetal growth and amniotic fluid volume (see box 12). ● Offer retinal assessment to women with pre-existing diabetes who did not have diabetic retinopathy at their first antenatal clinic visit. 40 weeks ● Offer tests of fetal well-being for women waiting for spontaneous labour. mode and management of birth – analgesia and anaesthesia (including anaesthetic assessment for women with comorbidities. 36 weeks ● Offer ultrasound monitoring of fetal growth and amniotic fluid volume. NICE clinical guideline 63 Quick reference guide 11 25 . ● Offer tests of fetal well-being for women waiting for spontaneous labour. 31 weeks ● No appointment (routine care offered to nulliparous women at 32 weeks). ● Offer investigations that would be offered to nulliparous women at 31 weeks in routine antenatal care. ● Offer information and advice about: – timing. such as obesity or autonomic neuropathy) – changes to hypoglycaemic therapy during and after birth – initial care of the baby – initiation of breastfeeding and the effect of breastfeeding on glycaemic control – contraception and follow-up.

Diabetes in pregnancy Diabetes in pregnancy Antenatal care Box 7 Blood glucose targets and monitoring ● Advise women to test fasting and 1-hour postprandial blood glucose levels after every meal during pregnancy. Agree individualised targets for self-monitoring. including support for a single failing organ system or postoperative care and those ‘stepping down’ from higher levels of care.9 mmol/litre and 1-hour postprandial blood glucose below 7. ● ● ● Advise: women to test their blood glucose before going to bed at night on the risks of hypoglycaemia and hypoglycaemia unawareness.5 and 5.8 mmol/litre. ● ● 5 Level 2 critical care is defined as care for patients requiring detailed observation or intervention. admit women immediately for level 2 critical care5. ● ● ● ● Box 8 Additional care for women taking insulin Offer: ● concentrated oral glucose solution to all women taking insulin ● ● glucagon to women with type 1 diabetes insulin pump therapy if glycaemic control using multiple injections is not adequate and the woman experiences significant disabling hypoglycaemia. For women with type 1 diabetes: offer ketone testing strips and advise women to test their ketone levels if they are hyperglycaemic or unwell exclude diabetic ketoacidosis as a matter of urgency in women who become unwell. Do not use HbA1c routinely in the second and third trimesters. The presence of diabetic retinopathy should not prevent rapid optimisation of glycaemic control in women with a high HbA1c in early pregnancy. especially in the first trimester women and their partners or family members on the use of oral glucose solutions and glucagon for hypoglycaemia. Box 9 Detecting and managing diabetic ketoacidosis If diabetic ketoacidosis is suspected during pregnancy. 12 NICE clinical guideline 63 Quick reference guide 26 . where both medical and obstetric care are available. Advise women to aim for a fasting blood glucose of between 3.

unless there is a risk of intrauterine growth restriction. Retinal assessment should be carried out by digital imaging with mydriasis using tropicamide. NICE clinical guideline 63 Quick reference guide 13 27 .Summary of recommendations and algorithm Diabetes in pregnancy Antenatal care Box 10 Retinal assessment for women with pre-existing diabetes Offer retinal assessment: ● as soon as possible after the first contact in pregnancy if it has not been performed in the past 12 months ● ● ● following the first antenatal clinic appointment at 28 weeks if the first assessment is normal at 16–20 weeks if any diabetic retinopathy is present. ● ● Do not offer: tests of fetal wellbeing before 38 weeks. Box 11 Renal assessment for women with pre-existing diabetes Offer: ● renal assessment at the first contact in pregnancy if it has not been performed in the past 12 months. Box 12 Monitoring and screening fetal development Offer: ● antenatal examination of the four-chamber view of the fetal heart and outflow tracts at 18–20 weeks ● ultrasound monitoring of fetal growth and amniotic fluid volume every 4 weeks from 28 to 36 weeks individualised monitoring of fetal wellbeing to women at risk of intrauterine growth restriction (those with macrovascular disease or nephropathy). Consider: referral to a nephrologist if serum creatinine is abnormal (120 micromol/litre or more) or total protein excretion exceeds 2 g/day thromboprophylaxis if proteinuria is above 5 g/day. ● ● Do not offer: ● eGFR during pregnancy.

3).2).1.0. 95% CI 0.8. smoking (OR 1.4).7 to 1.3).3.4 to 6.3 to 4.9. 95% CI 3.3 to 4.7) and severe hypoglycaemia (OR 1.9.9. 95% confidence interval (CI) 1.2 to 3.5 to 6.8.9). including miscarriage.9. Certain social and lifestyle factors have also been shown to be associated with poor pregnancy outcome:33 unplanned pregnancy (OR 1.7.9).2. 95% CI 2. fetal congenital anomaly and perinatal death.33 [EL = 3–4] Specific diabetes care risk factors. Asian or Other Ethnic Minority group) is not (OR 0.0.7 to 8.1 Preconception care Outcomes and risks for the woman and baby Description of the evidence No specific searches were undertaken for this section of the guideline.8) and sub-optimal glycaemic control before and during pregnancy were also associated with poor pregnancy outcome (preconception OR 3. However. no contraceptive use in the 12 months prior to pregnancy (OR 2.3) during pregnancy were not associated with poor pregnancy outcome.0).33 [EL = 3–4] This differs from the general maternity population in which both factors are associated with poor pregnancy outcomes.33 [EL = 3–4] However.2). recurrent hypoglycaemia (OR 1. 95% CI 0.6 to 1.32 [EL = 2+] Women with pre-existing complications of diabetes are more likely to have a poor pregnancy outcome (OR 2.7 to 2. 95% CI 0. How pregnancy affects diabetes Pregnancy can affect the control and complications of diabetes.1 to 5. 95% CI 2. 95% CI 2. 6 (intrapartum care) and 8 (postnatal care for women with diabetes).6.8) and sub-optimal approach of the woman to her diabetes during pregnancy (OR 3. 95% CI 0. Antenatal evidence of fetal growth restriction was associated with poor pregnancy outcome (OR 2.2.3 to 3.2 [EL = 3] Factors associated with poor pregnancy outcome (defined as a singleton baby with a major congenital anomaly born at any gestational age and/or a baby who died between 20 weeks of gestation and 28 days after birth) in women with type 1 and type 2 diabetes have been documented in the final report of the CEMACH diabetes in pregnancy programme.33 [EL = 3–4] These issues are discussed in Chapters 5 (diabetes care during pregnancy and antenatal care for women with diabetes). [EL = 3–4] The importance of planning pregnancy and the role of contraception is considered further in Section 3. but ethnicity (Black.0.1.1).2. 95% CI 0.2 to 7.0 to 4. 95% CI 2.2).3. 95% CI 1.4. A lack of local glycaemic control targets (OR 2.9.1 to 1. Diet and dietary supplements are considered further in Section 3.0 to 2.9 95% CI 1.3 3. maternity care risk factors and postnatal care risk factors were also identified in the CEMACH final report. 95% CI 1.5).0 to 3. but antenatal evidence of fetal macrosomia was not (OR 0. 95% CI 1. How diabetes affects pregnancy Women with type 1 and type 2 diabetes have an increased risk of adverse pregnancy outcomes. The evidence is drawn from publications identified in searches for other sections. There is increased frequency of hypoglycaemia and decreased hypoglycaemia awareness during pregnancy. 95% CI 1.5 to 1. a body mass index (BMI) of at least 30 kg/m² was not associated with poor pregnancy outcomes (OR 1. first trimester OR 3.9).33 [EL = 3–4] Maternal social deprivation is associated with poor pregnancy outcome for women with type 1 or type 2 diabetes (odds ratio (OR) 1.2. after first trimester OR 5. sub-optimal approach of the woman to her diabetes prior to pregnancy (OR 4.5 to 1. nephropathy (OR 2. 95% CI 1.3. Nausea and vomiting 28 . no preconception folic acid (OR 2. 95% CI 1.9) than women without.3 to 8. 95% CI 1.

7% in least-deprived quintile and 35. lack of discussion of diabetes-specific issues related to alcohol intake. antenatal evidence of fetal growth restriction. The following factors are associated with adverse pregnancy outcome: maternal social deprivation. this association was stronger and seen in women with type 1 diabetes (4. OR 0. unplanned pregnancy and lack of contraceptive use in the 12 months prior to pregnancy. Asian or Other Ethnic Minority origin compared to women with type 1 diabetes (48. pre-existing diabetes complications.4% in most-deprived quintile). This trend was not statistically significant in women of white ethnic origin with type 1 diabetes (18. For women of Black or Other Ethnic Minority origin.Preconception care in pregnancy can disrupt blood glucose control and severe nausea and vomiting (hyperemesis gravidarum) in women with diabetes can lead to ketoacidosis (see Section 5.6% in most-deprived quintile) and type 2 diabetes (3.4% Chinese and other ethnic background. Progression of retinopathy is more likely in women with more severe retinopathy. General anaesthesia in women with diabetes can also increase the risk of hypoglycaemia. implement and monitor policies to empower and support women with pre-existing 29 .3% White.1).5%).3% in least-deprived quintile and 21. lack of a baseline retinal examination in the 12 months before pregnancy and sub-optimal preconception care (excluding glycaemic control). increasing the frequency of hypoglycaemia and hypoglycaemia unawareness. Pakistani and Bangladeshi) and 3. women from ethnic minority groups are more likely to develop gestational diabetes (see Section 4.1% in most-deprived quintile). Existing guidance The standard set by the NSF for diabetes20 in relation to diabetes in pregnancy was for the NHS to develop. fetal congenital anomaly and perinatal death.3). 5.2 [EL = 3] Evidence statement Women with pre-existing type 1 and type 2 diabetes have an increased risk of adverse pregnancy outcomes. Asian or Other Ethnic Minority group.2). a much higher proportion of women with type 2 diabetes were of Black.33 [EL = 3–4] The CEMACH data for England showed that in women of white ethnic origin there was a clear increase in the number of women with type 2 diabetes with increasing quintile of social deprivation (6.8% Black (Black Caribbean. However. lack of discussion of fetal risks in diabetic pregnancy.9% in most-deprived quintile). including miscarriage. which reports 80.2 [EL = 3] The CEMACH data suggested that ethnicity is not associated with poorer pregnancy outcome for women with type 1 or type 2 diabetes (Black. sub-optimal approach of the woman to her diabetes before or during pregnancy and sub-optimal glycaemic control before or during pregnancy. The progression of certain complications of diabetes.8% in least and 45. Pregnancy is associated with progression of diabetic retinopathy. [EL = 3–4] Pregnancy can affect glycaemic control in women with diabetes. poor glycaemic control and hypertension (see Section 5.5 to 1. can be accelerated by pregnancy. Pregnancy may accelerate progression to end-stage renal disease in women with moderate to advanced diabetic nephropathy (see Section 5. and the risk of ketoacidosis. Black African and Black Other). They are also more likely to have unplanned pregnancies and less likely to have a measure of long-term glycaemic control in the 6 months before pregnancy (see Section 3.5% versus 8. specifically diabetic retinopathy and diabetic nephropathy. lack of preconception folic acid.5). 10.9.8). However.4% in least-deprived quintile and 59.4). smoking. Ethnicity The CEMACH descriptive study found that maternal ethnic origin of pregnant women with type 1 and type 2 diabetes (considered together) was not significantly different to the general maternity population of England. 95% CI 0.5). General anaesthesia in women with diabetes leads to a high risk of hypoglycaemia and a higher rate of Mendelson syndrome due to the higher resting gastric volume compared to women without diabetes (see Section 6.5% Asian (Indian. lack of local glycaemic control targets.

use of contraception and preconception counselling in women with type 1 and type 2 diabetes.2 The importance of planning pregnancy and the role of contraception Description of the evidence No specific searches were undertaken for this section of the guideline. which increases the likelihood of birth trauma. There were no research recommendations relating to the information that should be offered about outcomes and risks for the woman and the baby. It was suggested that keeping women with diabetes and their partners fully informed and involved in decision making would help to ensure that their experience of pregnancy and childbirth would be a positive one. Women with diabetes who are planning to become pregnant and their families should be offered information about how diabetes affects pregnancy and how pregnancy affects diabetes. the GDG’s recommendations are based on its consensus view of what information should be offered to women with pre-existing diabetes before pregnancy to support and explain its substantive recommendations regarding management options before. The NSF highlighted that maternity care for women with diabetes may be perceived as highly ‘medicalised’ with a tendency towards intervention in labour and birth and that this could make the experience negative or frightening.33 [EL = 3–4] 30 . The CEMACH diabetes in pregnancy programme provides data on current practice in England. during and after pregnancy. and that they should receive care that promotes their physical and psychological wellbeing and optimises the health of their babies. Wales and Northern Ireland in relation to planning pregnancy.Diabetes in pregnancy diabetes and women with gestational diabetes to optimise pregnancy outcomes. body weight and exercise the risks of hypoglycaemia and hypoglycaemia unawareness during pregnancy how nausea and vomiting in pregnancy can affect glycaemic control the increased risk of having a baby who is large for gestational age. The NSF stated that maternity care should ensure that all pregnant women have a positive experience of pregnancy and childbirth. stillbirth and neonatal death. Women with diabetes who are planning to become pregnant should be informed that establishing good glycaemic control before conception and continuing this throughout pregnancy will reduce the risk of miscarriage. advice and support that will help to reduce the risks of adverse pregnancy outcomes for mother and baby. From evidence to recommendations As no systematic searches were conducted for this section of the guideline. congenital malformation. 3. induction of labour and caesarean section the need for assessment of diabetic retinopathy before and during pregnancy the need for assessment of diabetic nephropathy before pregnancy the importance of maternal glycaemic control during labour and birth and early feeding of the baby in order to reduce the risk of neonatal hypoglycaemia the possibility of transient morbidity in the baby during the neonatal period. The evidence is drawn from publications identified in searches for other sections. which may require admission to the neonatal unit the risk of the baby developing obesity and/or diabetes in later life.20 Recommendations for outcomes and risks for the woman and baby Healthcare professionals should seek to empower women with diabetes to make the experience of pregnancy and childbirth a positive one by providing information. The information should cover: • • • • • • • • • the role of diet. both in terms of maternity care and management of diabetes during these periods and to reinforce the recommendations of the NSF for diabetes. It is important to explain that risks can be reduced but not eliminated.

0. lack of dietetic review before pregnancy (OR 1.0). lack of a baseline retinal examination in the 12 months before pregnancy (OR 2. The CEMACH enquiry found 51% (72/141) of the women with poor pregnancy outcome and 38% (55/144) of the women with good pregnancy outcome were documented as having not planned their last pregnancy (OR 1. 95% CI 0. retinopathy (OR 1.5) and assessment of sub-optimal preconception care (excluding glycaemic control.6 to 4.6 to 4.2.5. 95% CI 0.33 [EL = 3–4] However. 95% CI 0. 95% CI 1. Note that poor pregnancy outcome was defined as a singleton baby with a major congenital anomaly who gave birth at any gestation and/or a baby who died from 20 weeks of gestation up to 28 days after birth.8% of women with type 2 diabetes had pre-pregnancy counselling documented. 42% of women did not plan their last pregnancy.5.3) or assessment of albuminuria (OR 1.0 to 5.4).1.2 to 4.2.1). lack of a baseline test of renal function (OR 2.0).3). 95% CI 0.5). In the general maternity population.7.3).7. 95% CI 1.2% of women with type 1 diabetes and 24. 95% CI 1.33 [EL = 3–4] As noted in Section 3.1.5). OR 5.7 to 10.4 to 1.9) or possible caesarean section (OR 2.3 to 4.1 to 5.8 to 3.Preconception care The CEMACH descriptive study found 38. 95% CI 1. increased risk of induction of labour (OR 2. 95% CI 0. poor glycaemic control (OR 1.1. 95% CI 1.8. 95% CI 0.9.5 to 2.1 to 8.2 A pre-pregnancy glycaemic test in the 6 months before pregnancy was recorded for 40% of the women with type 1 diabetes and 29.2.3.9).5. 95% CI 0. Good pregnancy outcome was defined as a singleton baby without a congenital anomaly who survived to day 28 after birth.2.7).4. 95% CI 2. [EL = 3] The CEMACH enquiry reported that pre-pregnancy counselling included a discussion about: • glycaemic control in 51% of women with poor pregnancy outcome and 56% with good pregnancy outcome • diet in 42% of women with poor pregnancy outcome and 48% with good pregnancy outcome • contraception in 19% of women with poor pregnancy outcome and 36% with good pregnancy outcome • retinopathy in 30% of women with poor pregnancy outcome and 39% with good pregnancy outcome • nephropathy in 23% of women with poor pregnancy outcome and 25% with good pregnancy outcome • hypertension in 22% of women with poor pregnancy outcome and 28% with good pregnancy outcome • alcohol intake in 20% of women with poor pregnancy outcome and 25% with good pregnancy outcome • need for increased pregnancy surveillance in 54% of women with poor pregnancy outcome and 62% with good pregnancy outcome • fetal risks in 42% of women with poor pregnancy outcome and 58% with good pregnancy outcome • the increased chance of induction of labour for 32% of women with poor pregnancy outcome and 51% with good pregnancy outcome • the increased possibility of caesarean section for 39% of women with poor pregnancy outcome and 53% with good pregnancy outcome. 95% CI 0.8). social and lifestyle factors.8 to 4. such as unplanned pregnancy and lack of contraceptive use in the last 12 months.1).0 to 4.8 to 2.3.7 to 2.8.5 to 2. increased pregnancy surveillance (OR 1. 31 .4% of the women with type 2 diabetes.8.8). lack of discussion of fetal risks in diabetic pregnancy (OR 2. hypertension (OR 1. 95% CI 1. are associated with poor pregnancy outcome. lack of discussion of increased diabetes surveillance (OR 1.5).0 to 2. lack of discussion about diet (OR 1. 95% CI 1.33 [EL = 3–4] Sixty-six percent (71/108) of the women with poor pregnancy outcome and 45% (54/121) of the women with good pregnancy outcome were not documented as using any type of contraception in the 12 months before conception (OR 2. 95% CI 0.1).2). The CEMACH enquiry found an association between poor pregnancy outcome and a lack of discussion of diabetes-specific issues related to alcohol intake (OR 2. 95% CI 0.6 to 2. 95% CI 0.9 to 4. nephropathy (OR 0. the CEMACH enquiry found no association between poor pregnancy outcome and a lack of contraceptive advice before pregnancy (OR 1.

Women with diabetes who are planning to get pregnant are recommended to supplement their diet with folic acid until 12 weeks of gestation (see Section 3. Because of the importance of planning future pregnancies in women with diabetes.Diabetes in pregnancy The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that 38% (32/84) of women with type 2 diabetes and 40% (50/121) of women with type 1 diabetes were documented as having not planned their last pregnancy compared with 42% in the general maternity population. the chance of induction of labour and caesarean section have better pregnancy outcomes than women who do not have preconception care and advice.4). hypertension. P = 0. Since development of the fetal organs occurs during the first 3 months of pregnancy and good glycaemic control in the preconception period and the first trimester of pregnancy decreases the risk of congenital anomaly and miscarriage (see Section 3. Poor glycaemic control before pregnancy and in early pregnancy is associated with congenital malformations and miscarriage (see Section 3. P = 0. These medications should be discontinued before pregnancy to reduce the risk of congenital malformations and safe alternative medications should be prescribed if appropriate (see Section 3. and so good glycaemic control and avoidance of medications that could harm the developing fetus should be established before discontinuing contraception.35.3). diet. Some medications used to treat complications of diabetes can increase the risk of congenital malformation if used in the early stages of pregnancy. Contraceptive use in the 12 months prior to pregnancy was lower in women with type 2 diabetes (32%) compared with women with type 1 diabetes (59%.33 [EL = 3–4] Women with diabetes are at an increased risk of having a baby with a neural tube defect.2.33 Women with type 1 diabetes were more likely to have sub-optimal preconception glycaemic control (75%) than women with type 2 diabetes (60%) (P = 0. alcohol intake. angiotensin-II receptor blockers (ARBs) (also known as angiotensin-II receptor antagonists) used to treat hypertension and statins used to treat elevated cholesterol. [EL = 3–4] There are advantages to the woman and baby in planning pregnancy. women with diabetes are less likely to plan pregnancy than women without diabetes and young women with diabetes lack awareness of the importance of planning pregnancy and the role of preconception care (see Section 3. it is recommended that information and advice about contraception and the importance of planned pregnancy is offered to the women with diabetes in the postnatal period (see Section 8.36 planning pregnancy is particularly important for women with diabetes. fetal risks. retinopathy. P = 0.2). These medications include angiotensin-converting enzyme (ACE) inhibitors used to treat hypertension and nephropathy. [EL = 3–4] The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported there was no difference between the number of women with type 1 diabetes and those with type 2 diabetes who were documented as having commenced folic acid supplementation in the 12 months before pregnancy (45% [32/71] of women with type 2 diabetes and 49% [54/110] of women with type 1 diabetes.8).33 [EL = 3–4] The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that there was no difference between women with type 1 or type 2 diabetes with regard to whether they had a test for glycaemic control in the 12 months prior to pregnancy (83% versus 81%. Optimising glycaemic control before conception and in the first few weeks of pregnancy is of prime importance.4). The CEMACH enquiry found that a greater number of women with poor pregnancy outcomes (69%) compared with women with good pregnancy outcome (50%) were not documented as having commenced folic acid supplementation before pregnancy (adjusted OR 2.9).3 to 3.6).33 Only 32% (33/103) of women taking folic acid were on the high dose (5 mg/day). 32 . 95% CI 1.7). the need for increased pregnancy surveillance. contraception.013). nephropathy. However.001).66). Development of the fetal organs occurs in the first 3 months of pregnancy. Evidence statement Women with pre-existing diabetes who have preconception care and advice involving a discussion of glycaemic control.

Targeting postprandial hyperglycaemia is particularly important during pregnancy.4 and 5.4). Low glycaemic index (GI) diets appear to reduce postprandial hyperglycaemia and women on low GI diets have been reported to have babies with lower birthweights compared with women on high GI diets (see Sections 3. body weight and exercise Description of the evidence Diet The aims of dietary advice for women with diabetes who are planning a pregnancy are: • optimisation of glycaemic control.1). Women should be given information about the local arrangements for support.3 Diet. dietary supplements.1).3 and 5. stillbirth. avoiding large fluctuations of blood glucose. 33 . Dose Adjustment for Normal Eating (DAFNE) is an example of a structured education programme for people with type 1 diabetes in the UK (see Section 3. and neonatal hypoglycaemia and hypocalcaemia (see Sections 3.4).37 Hyperglycaemia in early pregnancy is associated with congenital malformations and miscarriage (see Section 3.1).4.4 and 5. the GDG’s recommendations are based on its consensus view of what information should be offered to women with diabetes in relation to planning pregnancy and the role of contraception to support and explain its substantive recommendations regarding management options before.Preconception care From evidence to recommendations As no systematic searches were conducted for this section of the guideline. 3. Recommendations for the importance of planning pregnancy and the role of contraception The importance of avoiding unplanned pregnancy should be an essential component of diabetes education from adolescence for women with diabetes. while avoiding ketosis and hypoglycaemia in women taking insulin • provision of sufficient energy and nutrients to allow normal fetal growth while avoiding accelerated fetal growth patterns. 4.  Diabetes Control and Complications Trial (DCCT)-aligned haemoglobin A1c (HbA1c) test. during and after pregnancy. medications for diabetes (including insulin regimens for insulin-treated diabetes) and medications for complications of diabetes will need to be reviewed before and during pregnancy • that additional time and effort is required to manage diabetes during pregnancy and that there will be frequent contact with healthcare professionals. especially postprandial blood glucose. including emergency contact numbers. glucose monitoring. Adjusting treatment to postprandial blood glucose levels is associated with better outcomes in women with type 1 diabetes or gestational diabetes than responding to fasting blood glucose (FBG) levels (see Sections 3. In later pregnancy it is implicated in accelerated fetal growth. There were no research recommendations relating to the information that should be offered about the importance of planning pregnancy and the role of contraception. Women with diabetes who are planning to become pregnant should be advised: • that the risks associated with pregnancies complicated by diabetes increase with the duration of diabetes • to use contraception until good glycaemic control (assessed by HbA1c) has been established • that glycaemic targets.

13) over and above that produced by high GI diets.3 to 3. 4 a.62 to 1.Diabetes in pregnancy A meta-analysis was identified that compared the effect on HbA1c of low GI diets with that of high GI diets.7) and large-for-gestational-age (LGA) babies (OR 1. The offspring of women with diabetes and coexisting 34 .33 [EL = 3–4] The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that 45% (32/71) of women with type 2 diabetes and 49% (54/110) of women with type 1 diabetes were documented as having commenced folic acid supplementation before pregnancy. 95% CI 1.4% more on the low GI diet than on the high GI diet (95% CI 8.6).6).33 [EL = 3–4] A case–control study compared folate metabolism in 31 pregnant women with diabetes to that in 54 pregnant women without diabetes.3. Not commencing folic acid supplements prior to pregnancy led to an increased risk of poor pregnancy outcome (OR 2. 95% CI 1.m. 2 a. the study found that. The meta-analysis found that low GI diets reduced HbA1c by 0.6) and preterm birth at less than 37 weeks of gestation (OR 5. pre-pregnancy BMI 30 kg/m² or more was independently associated with caesarean section (OR 3.4 to 18.m. [EL = 2+] Body weight Obesity is an independent risk factor for a number of adverse pregnancy outcomes including:37 • • • • • • • impaired glucose tolerance (IGT) hypertensive disorders caesarean section perinatal mortality macrosomia preterm birth congenital malformations. [EL = 2++] A prospective cohort study collected data on pre-pregnancy exposures and pregnancy outcome in 22 951 women (574 with diabetes.95. parity and obstetric history.m. [EL = 1++] A prospective randomised study40 was performed in 15 women with glucose intolerance diagnosed early in the third trimester of pregnancy. 95% CI 0.43 percentage points (95% CI 0.43 to 2.5. 95% CI 0.2.72 to 0.39 The meta-analysis included 14 studies and 356 nonpregnant people (203 with type 1 diabetes and 153 with type 2 diabetes).m. 95% CI 1. as compared with significantly lower in the insulin-treated group than the controls at 6 p.6).0).98.1 to 1..3). [EL = 2+] A cohort study included 196 women with pre-existing diabetes and 428 women with gestational diabetes.. when compared with pre-pregnancy BMI less than 20 kg/m². Taking HbA1c and fructosamine data together and adjusting for baseline differences.m. 95% CI 1. Weight gain during pregnancy was independently associated with hypertensive disorders of pregnancy (OR 1. and 6 a.8 to 6. and 8 p..m. The mean 2 hour neonatal plasma glucose concentration of the diet-treated group was significantly higher than that of other groups. The results showed that the mean plasma glucose concentrations of the diet-treated women were significantly greater than those of the controls (pregnant women with a normal glucose tolerance who ate according to appetite) at 10 a.2 to 1. 2 p.2). maternal age.m.42 There was no increased risk of major defects in the offspring of women without diabetes who were obese (relative risk (RR) 0.9).41 After controlling for type of diabetes.4.39 The study found no significant differences for any measures of folate metabolism. glycated proteins were reduced 7. but only 33 women were on the ‘high dose’ (5 mg) of folic acid.5) and no increased risk among women with diabetes who were not obese (RR 0. Dietary supplements Women with diabetes have an increased risk of having a baby with a neural tube defect (see Section 5. 95% CI 1.4 to 8.38. 1974 with a BMI 28 kg/m² or more). The CEMACH enquiry found 69% (83/120) of the women with poor pregnancy outcome and 50% (66/131) of the women with good pregnancy outcome were documented as not having commenced folic acid supplementation before pregnancy (this is similar to the general maternity population). [EL = 1−] Nausea and vomiting in pregnancy can disrupt blood glucose control (see Section 5.1.

A meta-analysis included six studies involving a total of 322 women with type 2 diabetes. The results showed that the gestational weight gain above Institute of Medicine guidelines was common and associated with multiple adverse neonatal outcomes.Preconception care obesity were three times as likely to have a major defect (RR 3. caesarean section. pregnancyinduced hypertension and LGA newborns. [EL = 2++] Exercise Moderate exercise has been found to improve blood glucose control in women with gestational diabetes (see Section 4. in pregnant women with diabetes on perinatal and maternal morbidity and mortality. [EL = 2+] A retrospective cohort study44 examined the relationship between gestational weight gain and adverse neonatal outcomes among term babies (37 weeks of gestation or more). [EL = 1+] Existing guidance The NICE antenatal care guideline9 recommends that healthy pregnant women (and those intending to become pregnant) should be informed that dietary supplementation with folic acid. The guidance recommends that women with diabetes who may become pregnant and those who are in the early stages of pregnancy should be prescribed 5 mg of folic acid per day. [EL = 2++] A prospective population-based cohort study43 of 1041 Latino mother–baby pairs assessed the combined influence of maternal weight and other anthropometric and metabolic characteristics on the birthweights of babies.1. HbA1c decreased more in the women in the dietary advice plus exercise group than in those in the dietary advice group alone. The results showed that there was an increased risk of adverse maternal and infant outcomes associated with excessive maternal weight. The review found no significant difference between exercise and the other regimens in any of the outcomes evaluated.46 All studies compared dietary advice with dietary advice plus exercise.3). The results showed that pre-pregnancy maternal BMI exhibited a much stronger influence than abnormal blood glucose tolerance on macrosomia. good outcomes were achieved without ketoacidosis. [EL = 1++] A Cochrane systematic review47 aimed to evaluate the effect of exercise programmes alone or in conjunction with other therapies such as diet. The recommended dose is 400 micrograms per day for the general maternity population.19 which aims to improve the nutrition of pregnant and breastfeeding mothers and children in low-income households. On average there was more weight loss in the dietary advice plus exercise groups. reduces the risk of having a baby with neural tube defects. No studies were identified in the population of pregnant women with pre-existing type 1 or type 2 diabetes.6) than those of women without diabetes. Obesity is a risk factor for a number of adverse perinatal outcomes. Evidence statement Low GI diets reduce postprandial glycaemia. 35 .4 to 1. 95% CI 1. NICE has developed public health guidance on maternal and child nutrition.5). suggesting that obesity and diabetes may act synergistically in the pathogenesis of congenital malformations. whereas women with a history of neural tube defects should take a higher dose (5 mg per day). An Expert Advisory Group report issued by the Department of Health48 recommended that women with no history of neural tube defects should take 400 micrograms per day.9% at 6 months (95% CI 0.3) and 1% at 12 months (95% CI 0.4 to 1. compared with no specific programme or with other therapies. In cohort studies of women with gestational diabetes who are obese undergoing moderate calorie restriction. weight gain and glucose intolerance.2 to 7. before conception and up to 12 weeks of gestation. [EL = 2−] A prospective cohort study45 evaluated the independent influence of pre-pregnancy BMI and glucose tolerance status on the presentation of diabetes-related adverse pregnancy outcomes in Spanish women. whereas gestational weight gain below guidelines was only associated with SGA babies. Dietary advice plus exercise was associated with a statistically significant mean decrease in HbA1c of 0.

Diabetes in pregnancy Exercise can help women with diabetes to lose weight and improve glycaemic control. All studies found a positive relationship between poor preconception glycaemic control and congenital malformations.org. Recommendations for diet. body weight and exercise Women with diabetes who are planning to become pregnant should be offered individualised dietary advice. However. preconception care means intensive treatment aimed at optimising blood glucose control prior to conception. body weight and exercise. 3. dietary supplements. changes in lifestyle. All women originally assigned to conventional treatment were changed to intensive treatment during pregnancy. Women with diabetes who are planning to become pregnant should be advised to take folic acid (5 mg/day) until 12 weeks of gestation to reduce the risk of having a baby with a neural tube defect. women with diabetes who are obese and intending to become pregnant should be advised of the risks to their own health and that of their babies and encouraged to start a supervised weight reduction diet. comorbidities. The non-randomised controlled trial49 compared 94 women originally assigned to intensive treatment (intensive treatment group) with 86 women originally assigned to conventional treatment (conventional treatment group). Folic acid is particularly important for women with diabetes planning a pregnancy because of the increased risk of congenital malformations. There were no research recommendations relating to the information that should be offered about diet. as for other women with increased risk of neural tube defects. 26 changed before conception 36 . when intending to become pregnant. Here. Women with diabetes have an increased risk of neural tube defects. From evidence to recommendations A number of factors interact during pregnancy to influence glycaemic control including physiological changes during pregnancy. insulin treatment and dietary factors. Women with diabetes who are planning to become pregnant and who have a body mass index above 27 kg/m² should be offered advice on how to lose weight in line with ‘Obesity: guidance on the prevention. identification. available from www. which include neural tube defects. Given the evidence that obesity is linked to a number of adverse perinatal outcomes.nice. Folic acid supplementation reduces the prevalence of neural tube defects.37 It follows that dietary advice should be given on an individual basis from someone with appropriate training and expertise. A non-randomised controlled trial and five cohort studies compared rates of congenital malformations in women attending for preconception care with those in non-attenders. uk/CG043. At present there is no evidence to suggest that these women would benefit from a larger dose than is recommended for women who do not have diabetes.4 Target ranges for blood glucose in the preconception period Description of the evidence Congenital malformations Fourteen studies were identified that examined the effect of preconception glycaemic control on congenital malformations. dietary supplements. the GDG’s view was that women with diabetes should take the higher dose of 5 mg per day. and there is no evidence that folic acid metabolism differs from that of women who do not have diabetes. assessment and management of overweight and obesity in adults and children’ (NICE clinical guideline 43).

4 > 14.1% ± 1.0%.0–9.6–13. There were 4.01). standard deviation (SD) 0. smoking and participation in preconception ‘counselling’. [EL = 2++] Nine studies examined the relationship between first-trimester HbA1c and the incidence of congenital malformations.3 ≥ 9.3 (1.8% (3/63) of congenital malformations in women with initial HbA1c less than 8.5) 3.05) and in the group with major malformations (9.2 Relative risk of congenital malformations in women with raised HbA1 compared with women with HbA1 of 9.3 9.9% (8/62) in women with initial HbA1c in the range 8.7) 1.7 (0.50–53 All five studies found significantly more congenital malformations in the nonattending group (P < 0. [EL = 2++] The five cohort studies compared women with type 1 diabetes attending a preconception clinic with non-attenders.Preconception care and 60 changed after conception.6%. 95% CI 1. In all cases the first-trimester HbA1c was significantly higher in the non-attending group (P < 0.6 5.7% in the conventional treatment group (P = 0.7) 3.9–5. [EL = 2++] Table 3.5) 2.3 (1.3% (6/17) in women with HbA1c 10.3%.9% and 35. Four of these studies reported first-trimester HbA1c levels.9%.3% or less55 First-trimester HbA1 (%) ≤ 9.2–7.8. [EL = 2++] A cohort study of 303 pregnant women with type 1 diabetes55 found the rate of major malformations increased significantly in women whose first-trimester HbA1 values exceeded 12.3) 12. when compared with women without diabetes.6–6. duration of diabetes.3% versus 2.001) than in the group without malformations (8. [EL = 2++] Table 3.4 (0. parity.4) In a cohort study of 142 women with type 1 diabetes56 17/142 pregnancies were complicated by malformation (six minor and 11 major).1 Relative risk of congenital malformations for different gradients of first-trimester HbA1c for women with type 1 diabetes compared with women without diabetes35 HbA1c at less than 14 weeks of gestation (%) Congenital malformations < 5.02 after adjusting for White’s classification of diabetes.7% (normal mean 5.9.9%) 4/10 (40%) RR (95% CI) 1.6) 4.9–8.3%) 7/18 (38.57) or 12 SD above the mean (see Table 3.3–9.0–5.6 (0. Nine congenital malformations were identified.9) A study of 435 pregnant women54 (289 with type 1 diabetes and 146 with type 2 diabetes) compared women with first-trimester HbA1c more than 8% with women with first-trimester HbA1c less than 8%.1.3–8.0 11. OR 3.4 1/47 7/170 8/252 6/133 4/61 RR (95% CI) 1. age at onset of diabetes.7–35. The mean HbA1c at conception in the intensive treatment group was 7.3–8. 12.8 (1.4–11.3 to 8.2%) 2/46 (4.2).06).2 (4. P < 0. SD 1.8–14.4–1. Women with first-trimester HbA1c more than 8% had higher rates of congenital malformations (8. SD 1. no description of preconception counselling reported).4% ± 1.5.4).0) 13.8 6.8 (4.1–12.7 mmol/litre or less. The goals for all pregnant women were fasting glucose level 3.0 1.36.7 12.05).9.0001).3% compared with 8.5 mmol/litre and 1 hour postprandial glucose level 7.0% or more. The RR. The mean initial HbA1c was significantly higher in the group with minor malformations (9.05).1–9.4 Congenital malformations 3/99 (3%) 4/77 (5.0%.5%. SD 1. eight in the conventional treatment group (P = 0. [EL = 2++] 37 . A cohort study of 691 pregnancies and 709 babies of 488 women with type 1 diabetes35 found first-trimester HbA1c to be associated with occurrence of malformations (P = 0.3–40. P < 0. P < 0. for different gradients of firsttrimester HbA1c are presented in Table 3.0 8.0 (1.

9% and 21.1% (1.2% (normal non-diabetic mean 6%. P = 0.2 SD above the normal mean. The rate of miscarriage was significantly lower in the preconception care group (7% versus 24%.5 mmol/litre and 1 hour postprandial glucose level 7. The normal non-diabetic mean (SD) HbA1 in the laboratory used in the study was 7. Within the normal range.7% in the conventional treatment group (P = 0.1% (2/39) of women with HbA1c in the range 7. [EL = 2++] 38 . loss rates in women with diabetes increased in an approximately linear fashion with increasing levels of HbA1c (y = 0. There were nine congenital malformations.02).0–8. All women originally assigned to conventional treatment were changed to intensive treatment during pregnancy.7x − 19. P = 0.05).7% (5/23) of women with HbA1c 10.0–6. [EL = 2++] In a cohort study of 229 women with type 1 diabetes.9% (8/35) of women with HbA1c in the range 8.05).9%. Fasting (P = 0.1% enrolled less than 21 days after conception.1. P < 0.5%.4% [14]).002) and 14 weeks of gestation (P = 0. [EL = 2++] Miscarriage Eight studies were identified that examined the effect of preconception glycaemic control on miscarriage. One non-randomised trial and two cohort studies compared rates of miscarriage in women who received preconception care with rates in women with no preconception care.0008) and at 9 weeks of gestation (P = 0.7 mmol/litre or less. The goals for all pregnant women were fasting glucose level 3. The normal non-diabetic mean (SD) HbA1c was 5. There was no significant difference between the rates of miscarriage in the intensive and conventional treatment groups (13. The threshold for increased risk of malformation was a preprandial glucose concentration of 6.005. the remaining 24.1% ± 1.5%. Ten of the 14 malformations occurred in 42 women whose levels exceeded this value.5% and six were in women whose HbA1c was more than 13.4% ± 1.5%.6. [EL = 2++] A cohort study of 99 pregnant women with type 1 diabetes62 compared outcomes in 28 women who had attended a preconception clinic with those in 71 women who had enrolled after conception.5 ± 1.004) blood glucose levels were significantly higher in women who miscarried and were significant predictors of loss when analysed by logistic regression. Seven of the nine malformations were in women whose HbA1c was more than 11. Above the normal range. The mean HbA1c at conception in the intensive treatment group was 7.0001). there was no relationship between the level of HbA1c and the rate of pregnancy loss in women with diabetes or without diabetes.85–5. The non-randomised controlled trial49 compared 94 women originally assigned to intensive treatment (intensive treatment group) with 86 women originally assigned to conventional treatment (conventional treatment group).01). HbA1c at the first antenatal visit and at 9 weeks of and 14 weeks of gestation were significantly higher in women who had miscarriages (P = 0. 26 changed before conception and 60 changed after conception.Diabetes in pregnancy A cohort study of 116 women with type 1 diabetes57 found HbA1c was significantly higher in the group with major congenital malformations (13) than in the group without (9. Sixteen percent (62/386) of women with diabetes had miscarriages as did 16% (70/432) of women without diabetes.04). HbA1c levels were significantly lower in the preconception care group at the first antenatal visit (P = 0. [EL = 2++] A cohort study of 105 women with type 1 diabetes58 found a significant increase in the incidence of malformation above a first-trimester HbA1a+b+c of 9.1% (14) of women had babies with congenital malformations. compared with four malformations in the 45 women with lower values. P = 0.02).0% (0. range 5.8%). [EL = 2++] A cohort study of 83 pregnant women59 (63 with type 1 diabetes and 20 with type 2 diabetes) examined the effect of first-trimester HbA1c on congenital malformations.1%). P = 0.015).3% [18] versus 10.0 versus 8.01) and postprandial (P = 0.61 The percentage of women who enrolled before conception was 75. [EL = 2++] A cohort study compared pregnancy outcomes in relation to glycaemic control in 386 women with type 1 diabetes and 432 women without diabetes. There were no malformations (0/19) in women with HbA1c 6.0% or more. 22.3% compared with 8.5 ± 1.9% or less compared with malformations in 5.02.6 mmol/litre or an initial HbA1 of 13% (6.9%.60 6.6–9. all occurring in women with first-trimester HbA1c more than 9. P < 0. P = 0.

5%.1%).0% (normal non-diabetic mean (SD) 5.15 mmol/litre or an initial HbA1 of 12% (6.4%.1) in the adverse outcome group compared with 7. [EL = 2++] A study of 57 births in women with type 1 diabetes examined effect of glycaemic control at different times before and during pregnancy on birthweight. The rate of miscarriage may be underestimated in this study. PPV 0. [EL = 2++] The remaining five studies examined the effect of first-trimester HbA1c on miscarriage. 1/22 occurred in women whose HbA1c was 7.1% (1.67 Multiple regression found a correlation with HbA1c at 0–12 weeks of gestation (multiple r = 0. NPV 0.89.002) and to measure glucose daily at conception (23% versus 35%. The median HbA1c 0–3 months before conception was 8. Some women may experience miscarriage before enrolling for antenatal care. [EL = 2++] A cohort study of 303 women with type 1 diabetes55 found miscarriage rates were significantly increased when first-trimester HbA1 was greater than 11. The normal non-diabetic HbA1c level was 5. HbA1c more than 15. This study considered miscarriages and congenital malformation together and calculated predictive values for adverse outcome as follows: HbA1c more than 7.001). P = 0. specificity 14%. The mean firsttrimester HbA1c was 9.9% ± 2.0% (0. which was not significantly different from the mean HbA1c values of those pregnancies resulting in a malformation (10.5%. [EL = 2++] A cohort study of 83 women59 included 63 women with type 1 diabetes and 20 with type 2 diabetes.5%).6) than women whose pregnancies progressed beyond 20 weeks (10. A cohort study of 105 women with type 1 diabetes58 reported miscarriage in 18 women.8%). No miscarriages occurred in women with firsttrimester HbA1c less than 7.5–9. Miscarriages occurred in 22 pregnancies (26. P < 0. sensitivity 97%. sensitivity 100%. [EL = 2+] 39 .8–8.9%) or no adverse outcome (8.48. F = 0.5%. The range of gestational age at entry was 6–16 completed weeks.5%.00005).001). The threshold for increased risk of miscarriage rate was a preprandial glucose concentration of 7. The initial glucose and HbA1c values of women who had no preconception care and who had a miscarriage were significantly higher compared with women who received preconception care and whose pregnancies progressed beyond 22 weeks of gestation (P < 0.05). The normal non-diabetic mean (SD) HbA1c was 5.2 SD above the normal mean. The findings suggest that poor metabolic control around conception.05).3–9.5%.Preconception care A cohort study of 94 women with type 1 diabetes63 compared 59 women who attended a preconception clinic with 35 non-attenders. is the important factor. P < 0.41. P = 0.05).1% (1.85. Women in the adverse outcome group were also less likely to have preconception guidance (42% versus 59%.005). The normal non-diabetic mean (SD) HbA1 was 7. PPV 0.3%). PPV 1.1%).3%.0 (interquartile range 7. HbA1c more than 9. The normal non-diabetic range of HbA1c was 5.4% ± 2. 21/22 occurred in women whose HbA1c was more than 11.5% and 19/22 occurred in women whose HbA1c was more than 13. Women who had a miscarriage had significantly higher HbA1 (12. [EL = 2++] A cohort study of 84 pregnancies in 68 women with type 1 diabetes64 measured HbA1 (normal non-diabetic range 5.74.0 ± 0.3. There was no association with glycosylated proteins or glycosylated albumin.3 ± 1. [EL = 2++] In a cohort study of 215 women with type 1 diabetes.5%.5–8. [EL = 2++] Other neonatal outcomes A cohort study of 1218 pregnancies in 990 women66 considered the effect of glycaemic control at different times before and during pregnancy on adverse outcomes (perinatal death and/or congenital malformation). [EL = 2++] A cohort study of 116 pregnancies in 75 women with type 1 diabetes65 found that miscarriages were significantly more likely to occur when HbA1 was 12% or more (P < 0. PPV 0.60 52 women (24%) had miscarriages.5%). specificity 82%.5–8.77.9% (0.019).5%. NPV 0.57%)). glycosylated proteins and glycosylated albumin at 9 weeks of gestation. [EL = 2++] A cohort study of 83 women59 included 63 women with type 1 diabetes and 20 with type 2 diabetes. Miscarriage occurred in 5/59 women who attended the preconception clinic and in 10/35 non-attenders (P < 0. specificity 94%. rather than in the weeks immediately preceding the miscarriage.5) in the other group (P = 0.7 ± 0. HbA1c more than 13. sensitivity 55%. NPV 1.6 (6.

89. It is important to note that the data for these two groups of women were collected in different ways. 91% (194/213) used HbA1c to monitor glycaemic control. [EL = 2++] A UK RCT69 evaluated a course teaching flexible intensive insulin treatment. There was evidence that targets for glycaemic control had been set before pregnancy for only 28% of 369 women.4% as compared with 9. five episodes of cerebral oedema.0001).5% versus 8. The main concerns of the enquiry panels were a lack of timely clinical input by heath professionals to improve glycaemic control and that insulin regimens advised were inadequate to achieve control. Women who had sub-optimal preconception glycaemic control were more likely to go on to have a poor pregnancy outcome (OR 3. P < 0. Hypoglycaemia was significantly more common in women in the preconception care group (26. [EL = 2++] A cohort study of 84 pregnant women with type1 diabetes68 recruited before 9 weeks of gestation found hypoglycaemia requiring third-party assistance occurred in 71% of women.32 Of the women in the CEMACH descriptive study of 3808 diabetic pregnancies who had a HbA1c measurement before conception. Wales and Northern Ireland.001). Consequences of maternal hypoglycaemia included several grand mal seizures. A cohort study of 239 women with type 1 diabetes51 compared 143 women who attended a prepregnancy clinic with 96 women managed over the same period who had not received pre-pregnancy care.3%. namely.15 to 7. The course provided skills to enable adults with type 1 diabetes to replace insulin by matching it to a desired carbohydrate intake on a meal-by-meal basis. The HbA1c at 6 months in adults who attended a 5 day training course (immediate DAFNE group) was 8. postprandial glucose or glycohaemoglobin concentrations. Sixteen percent of the 379 women did not have a glycaemic control test in the year before pregnancy and for more than a quarter (27%. two road traffic accidents and comminuted fracture of the tibia and fibula. Although all women in the study were undergoing intensive insulin therapy there was no association between hypoglycaemia and glycaemic control as measured by preprandial glucose. One of the 12 women who had a baby with a malformation had experienced hypoglycaemia in the first 9 weeks of pregnancy. The first-trimester mean HbA1c was significantly lower in the preconception care group (8. 101/379) of women there was no documentation available. 6% (13/213) used HbA1.4% versus 10. 28% had a value below 7%.1. There was no association between hypoglycaemia and early pregnancy loss or major malformations in babies.33 [EL 3–4] This compares with 37% of 3808 women with diabetes in the full cohort having had a test of glycaemic control in the 6 months before pregnancy. and 1% (3/213) used fructosamine as the only test. The 2002 CEMACH enquiry found that of units expected to provide maternity care for women with diabetes in England. χ²=11. 95% CI 2.5%. adjusted for maternal age and deprivation). weight or lipids.4% in those who received usual care for 6 months and then attended a training course (delayed DAFNE group). For 52% of women there was no documentation in the medical records about glycaemic control targets. Only 74% (158/213) of the units used an HbA1c assay aligned with the Diabetes Control and Complications Trial (DCCT). with a peak incidence at 10–15 weeks. It was assessed that 79% of 354 women with available documentation had sub-optimal glycaemic control before pregnancy. One other woman required second-party assistance by administration of juice in the home. There was a significant improvement in general wellbeing and treatment satisfaction in the immediate DAFNE group but no difference in severe hypoglycaemia. [EL = 1+] Current practice Fifty-eight percent (218/379) of women in the CEMACH diabetes in pregnancy enquiry were reported to have had a test of glycaemic control in the 12 months before pregnancy. [EL = 2++] A cohort study of 84 women recruited prior to conception and 110 women referred at 6–8 weeks of gestation36 found that at 2–8 weeks of gestation in the preconception group 42% recorded no hypoglycaemic episodes. P < 0.02. The other 45 women who had severe hypoglycaemia had no abnormal outcomes. 35% recorded one or two episodes per week and 23% experienced several episodes per week. One woman experienced loss of consciousness requiring a glucagon injection and intravenous glucose therapy in an emergency department. [EL = 3] 40 . The impact of diabetes on ‘freedom to eat as I wish’ significantly improved in the immediate DAFNE group as compared with the delayed group.Diabetes in pregnancy Maternal hypoglycaemia Three studies were identified that reported maternal hypoglycaemia as an outcome. either alone or in combination with other tests. from review of the medical records and from questionnaires sent to health professionals involved in their care.

There is a substantial increase in the incidence of miscarriages and congenital malformations in women whose first-trimester DCCT-aligned HbA1c exceeds 10%. Good glycaemic control before conception puts the woman at an increased risk of hypoglycaemia. Women with diabetes whose HbA1c is above 10% should be strongly advised to avoid pregnancy. Therefore women are advised to aim to keep their HbA1c within the normal non-diabetic range.1%. Targets for blood glucose monitoring should. Very high levels of HbA1c (more than 10%) substantially increase the risk of adverse outcomes and it is the GDG’s view that pregnancy should be avoided when DCCT-aligned HBA1c exceeds 10%. The DCCT compared intensive treatment with conventional treatment where intensive treatment was a package of care that included self-monitoring of blood glucose at least four times a day and a monthly measurement of HbA1c. If it is safely achievable. From evidence to recommendations Evidence shows that the risk of congenital malformations increases with increasing values of HbA1c above the normal (non-diabetic) range. Women should be reassured that any reduction in HbA1c towards the target of 6.0%.70 The DCCT involved 1441 people with type 1 diabetes.1% is likely to reduce the risk of congenital malformations. Recommendations for target ranges for blood glucose in the preconception period Individualised targets for self-monitoring of blood glucose should be agreed with women who have diabetes and are planning to become pregnant.1%.5 Monitoring blood glucose and ketones in the preconception period Description of the evidence No studies were identified that assessed how blood glucose or ketones should be monitored in the preconception period. where HbA1c is aligned with the DCCT this is below 6. 3. therefore. and women may be attempting to achieve optimal glycaemic control for several months until conception occurs. taking into account the risk of hypoglycaemia. The aim should be for the woman to maintain her HbA1c below 7. women with diabetes who are planning to become pregnant should aim to maintain their HbA1c below 6. The goal of intensive therapy was blood glucose concentrations as close to the non-diabetic range 41 .Preconception care Existing guidance The NSF for diabetes20 recommends ‘Women should be encouraged and supported to monitor their blood glucose regularly. the risk of miscarriage and congenital malformations increases as first-trimester HbA1c increases above the normal range.’ Evidence statement A large number of high-quality cohort studies have established an association between blood glucose control before conception and the incidence of miscarriage and congenital malformations. Women should be reassured that any reduction in HbA1c towards the target of 6. and to adjust their insulin dosage in order to maintain their blood glucose levels within the normal (non-diabetic) range.1% is likely to reduce the risk of congenital malformations. Threshold values for increased risk of miscarriage and congenital malformations are difficult to identify because of differences between studies in measures of glycosylated haemoglobin and the normal non-diabetic range. In general. be agreed with the individual taking into account the risk of hypoglycaemia. There were no research recommendations relating to target ranges for blood glucose in the preconception period.

2). the aim being to avoid hypoglycaemia. Women with diabetes who are planning to become pregnant should be offered a meter for self-monitoring of blood glucose. 42 . An ancillary study compared pregnancy outcomes in 94 women originally assigned to intensive treatment with 86 women originally assigned to conventional treatment. The mean ± SD value for all glucose profiles in the intensive therapy group was 8.and postprandial testing will give a more accurate picture of the blood glucose profile than preprandial testing alone (see Sections 3.3% compared with 8.2).7% in the conventional treatment group (P = 0. 26 changed before conception and 60 changed after conception. Women with type 1 diabetes who are planning to become pregnant should be offered ketone testing strips and advised to test for ketonuria or ketonaemia if they become hyperglycaemic or unwell. No studies were identified on monitoring for ketones in the preconception period. Nine congenital malformations were identified. Therefore.1 and 5. [EL = 2++] Evidence statement No studies were identified on how to monitor blood glucose and ketones in the preconception period. [EL = 1++] The DCCT protocol required that women in the conventional treatment group change to intensive therapy while attempting to become pregnant and during pregnancy.001). Therefore the GDG recommends that women should be advised in the use of ketone strips as part of preconception care. HbA1c was maintained at a significantly lower level in the intensive treatment group compared with the conventional treatment group (P < 0. women with insulin-treated diabetes should be offered meters to facilitate frequent monitoring alongside intensified insulin regimens.0001). Recommendations for monitoring blood glucose and ketones in the preconception period Women with diabetes who are planning to become pregnant should be offered monthly measurement of HbA1c.1 mmol/ litre in the conventional therapy group (P < 0. In the absence of any evidence for the effectiveness of self-monitoring of ketones in the preconception period.1% ± 1. In pregnancy there is accelerated ketosis (see Section 5. An RCT found that a package of care that included self-monitoring of blood glucose at least four times a day and a monthly measurement of HbA1c improved blood glucose control in people with type 1 diabetes and reduced the progression of complications of diabetes. Women with diabetes who are planning to become pregnant and who require intensification of hypoglycaemic therapy should be advised to increase the frequency of self-monitoring of blood glucose to include fasting and a mixture of pre.4. Evidence of the effectiveness of self-management of blood glucose using meters supports the provision of meters (see Section 5. A mixture of pre. the GDG’s view is that recommendations should be based on best current practice.6 ± 1.3). An ancillary study found that the incidence of congenital malformations in the babies of women with type 1 diabetes was also reduced. an approach to blood glucose monitoring based on that included in the intensive treatment package from the DCCT is recommended (self-monitoring of blood glucose at least four times a day and a monthly measurement of HbA1c).and postprandial levels. 5.4% ± 1.0001). The mean HbA1c at conception in the intensive treatment group was 7. eight of which were in the conventional treatment group (P = 0.49 All women originally assigned to conventional treatment were changed to intensive treatment during pregnancy.06). From evidence to recommendations In the absence of direct evidence for the effectiveness of self-monitoring of blood glucose in the preconception period.Diabetes in pregnancy as possible.8 ± 3. The study found intensive therapy delayed the onset of complications of diabetes and slowed their progression. The GDG’s view is that the same monitoring regimen should be used in the preconception period and during pregnancy.7 mmol/litre compared with 12.

Six studies were rated as of ‘poor’ quality. There were no significant differences between the glibenclamide and insulin groups in the percentage of babies who were LGA (12% versus 13%). three case series and one case– control study. had lung complications (8% versus 6%) or neonatal hypoglycaemia (9% versus 6%). Approximately 83% were Hispanic. glimepiride. increased incidence of pre-eclampsia or other adverse maternal or neonatal outcomes. three retrospective cohort studies. were admitted to a NICU (6% versus 7%).99).0 mmol/litre in the insulin group (P = 0. [EL = 1++] 43 . glipizide. There was no significant difference in the rate of major malformations between those exposed to oral hypoglycaemic agents and those not exposed (10 studies. Two other medications are used to stimulate insulin release (nateglinide and repaglinide). α-glucosidase inhibitors (acarbose) and thiazolidinediones (pioglitazone and rosiglitazone).9 ± 0. The meta-analysis included ten studies which reported on 471 women exposed to oral hypoglycaemic agents in the first trimester and 1344 women not exposed. gliclazide. 12% were white and 5% were black.Preconception care There were no research recommendations relating to monitoring blood glucose and ketones in the preconception period. macrosomic (7% versus 4%).73 The review reported that metformin use during the first trimester of pregnancy was not associated with an increased risk of major malformations. Cord-serum insulin concentrations were similar in the two groups and glibenclamide was not detected in the cord serum of any baby in the glibenclamide group.9 ± 1. tolbutamide (six studies). The mean blood glucose concentrations during treatment were 5. There are four main categories of oral hypoglycaemic agents: sulphonylureas (chlorpropamide. The oral hypoglycaemic agents used in the studies included chlorpropamide (eight studies). 95% CI 0.9 mmol/litre in the glibenclamide group and 5. glibenclamide (also known as glyburide). Eight women (4%) in the glibenclamide group required insulin treatment. glibenclamide (four studies).6 The safety of medications for diabetes before and during pregnancy The safety of diabetes medications (including oral hypoglycaemic agents and insulin preparations) for use during pregnancy is considered in this section. [EL = 2++] A further systematic review included seven additional studies that looked at the use of metformin in the first trimester of pregnancy. Description of the evidence Oral hypoglycaemic agents Oral hypoglycaemic agents are used to maintain blood glucose control in people with type 2 diabetes. metformin (five studies) and phenformin (three studies).3). A systematic review and meta-analysis examined the relationship between first-trimester exposure to oral hypoglycaemic agents and subsequent congenital abnormalities and neonatal mortality. Most women in the studies had type 2 diabetes. biguanides (metformin). six in women with polycystic ovary syndrome and one in women with type 2 diabetes. 3. two were ‘fair’ and two were ‘good’. The use of metformin during pregnancy in women with polycystic ovary syndrome is associated with a reduction in miscarriage in early pregnancy.74 The study involved 404 women with gestational diabetes. There were three prospective cohort studies. Each works in a different way and is suitable for different clinical situations. OR 1. or had fetal anomalies (2% and 2%). Women with type 1 diabetes and gestational diabetes were also present in some studies. [EL = 2+] An RCT compared glibenclamide with insulin for the treatment of gestational diabetes. The effectiveness of insulin preparations and regimens (including continuous subcutaneous insulin infusion (CSII) pumps) is considered in Section 5.65 to 1.05.71 The review sought to account for the potential confounding effect of maternal glycaemic control. There were no reports of birth defects. a reduction in fasting serum insulin levels and in the incidence of gestational diabetes. gliquidone and tolbutamide).70). Women were randomly assigned between 11 and 33 weeks of gestation to receive either glibenclamide (n = 201) or insulin (n = 203). weight loss.72 All the studies were small observational studies. [EL = 2+] Another systematic review evaluated the safety of metformin in pregnancy for women with diabetes and polycystic ovary syndrome.

756. The observational studies identified congenital malformations in some babies of women taking metformin. Glipizide has been shown to cross the placenta. The reference guide noted that high maternal blood glucose is associated with maternal and fetal effects. An observational study that looked at the incidence of congenital malformation in women with type 2 diabetes found no association between oral hypoglycaemic agents and organogenesis or congenital malformations. [EL = 2+] A reference guide to drugs in pregnancy and lactation reported that there were limited data on the use of metformin. Small amounts of glibenclamide may cross the placenta. Gliclazide and gliquidone were not reviewed.220. glimepiride. Conversion from oral hypoglycaemic agents to insulin was protective for perinatal mortality compared with oral hypoglycaemic agents alone (OR 0. but that insulin is still the treatment of choice. The observational studies identified congenital malformations in some babies of women taking acarbose. Metformin has been shown to cross the placenta.77 [EL = 3] The reference guide stated that one case report had investigated the use of nateglinide in a pregnant woman with diabetes. acarbose. glipizide and glibenclamide in pregnant women and suggested they present a low risk to the fetus. but it was suggested that they present a moderate risk to the fetus. The oral hypoglycaemic agents assessed were metformin and glibenclamide. but the reference guide suggested that it may present a risk of skeletal deformities. growth retardation and intrauterine death. but the rate was not compared with the expected rate of congenital malformations in babies born to women with diabetes who were not taking metformin. 95% CI 0. and that insulin is the treatment of choice to prevent hyperglycaemia. including congenital abnormalities. whereas perinatal mortality rates (per 1000 births) were higher in the group that used oral hypoglycaemic agents alone (125. No data were found on the use of repaglinide or pioglitazone in pregnant women.76 The women were subdivided into: oral hypoglycaemic agents alone (n = 93 pregnancies).024). Fetal anomaly rates were similar across the three groups. P = 0. Evidence suggested that chlorpropamide and tolbutamide present a risk to the fetus if taken by women in the third trimester of pregnancy. but it was suggested that it presents a risk to the fetus.061 to 0.77 [EL = 3] The reference guide reported that no studies had investigated the use of glimepiride in pregnant women.Diabetes in pregnancy An RCT of metformin for the treatment of gestational diabetes (the Metformin in Gestational Diabetes (MIG) trial) is due to report soon.77 [EL = 3] The reference guide reported that three observational studies had investigated the use of acarbose in pregnant women with diabetes.77 [EL = 3] The reference guide reported that case reports had investigated the use of glipizide in pregnant women. and the reference guide suggested that it presents a low risk of developmental toxicity.75 [EL = 1−] A retrospective cohort study examined the effects of oral hypoglycaemic agents in 379 singleton pregnancies of women with type 2 diabetes.77 [EL = 3] The reference guide reported that 14 small observational studies had investigated the use of metformin in pregnant women with diabetes. It is not known whether glimepiride can cross the placenta. and the rate was not compared with the expected rate of congenital malformations in babies born to women with diabetes who were not taking acarbose.77 [EL = 3] 44 . The reference guide suggested that use of metformin may decrease fetal and infant morbidity and mortality in developing countries where the proper use of insulin is problematic. nateglinide. The observational studies suggested that use of glibenclamide may decrease fetal and infant morbidity and mortality in developing countries where the proper use of insulin is problematic. It is not known whether nateglinide can cross the placenta. and the reference guide suggested that it may be an acceptable alternative to insulin for women with gestational diabetes. There were no differences in perinatal outcomes.77 [EL = 3] The reference guide reported that several observational studies had investigated the use of glibenclamide in pregnant women. A pilot study randomised 14 women to insulin and 16 to metformin. P = 0.003). and insulin alone or converted from diet alone to insulin (n = 37). The data suggest that metformin and glibenclamide are not teratogenic. No comparative studies were found on the use of rosiglitazone in pregnant women. converted from oral hypoglycaemic agents to insulin (n = 249).

aspart and glulisine and the long-acting analogues detemir and glargine. Ten case reports relating to tolbutamide use identified no increase in the rate of congenital abnormalities over and above that expected in women with diabetes. There was no significant difference in mean reduction in HbA1c over the 45 . pioglitazone and rosiglitazone advise pregnant women to avoid them. It is not known whether repaglinide can cross the placenta.72. No differences were observed in 24 hour mean plasma glucose profiles or HbA1c values. symptomatic neonatal hypoglycaemia and one baby experienced prolonged neonatal hypoglycaemia and seizures.77 [EL = 3] Insulin analogues Insulin analogues are synthetic insulins created by modifying the chemical structure of insulin to produce either faster acting preprandial insulin or longer acting basal insulin.77 [EL = 3] The reference guide stated that one case report had investigated the use of rosiglitazone in a pregnant woman with diabetes. and insulin is normally substituted in women with diabetes.78 The manufacturers of nateglinide.79 There were fewer episodes of nocturnal hypoglycaemia (RR 0. Insulin Insulin has been shown to be compatible with pregnancy and is recommended as the drug of choice for pregnant women with diabetes. 95% CI 0. There is information on the use of oral hypoglycaemic agents during breastfeeding in Section 8.48.10) and severe hypoglycaemia (RR 0. 95% CI 0. It is not known whether pioglitazone can cross the placenta. P = 0. but the reference guide suggested that it may result in post-implantation losses.1. Progression of retinopathy and perinatal complications were similar in both groups. four babies of women who had taken chlorpropamide experienced prolonged.20 to 1. delayed development and reduced fetal weight. If oral hypoglycaemic drugs are used then therapy should be stopped at least 2 days before birth.Preconception care The reference guide reported that no studies had investigated the use of repaglinide in pregnant women. Out of 74 case reports identified. P = 0. No studies were identified in relation to the effectiveness and safety of insulin glulisine or insulin detemir in pregnancy.46. and the reference guide suggested that it may affect fetal growth.36 to 1.14. One case of prolonged hypoglycaemia in a baby following maternal treatment with tolbutamide was identified. acarbose and repaglinide should be avoided during pregnancy and are normally substituted with insulin. but the reference guide suggested that it may present a risk of placental toxicity. Insulin aspart: An RCT compared 157 pregnant women with type 1 diabetes treated using insulin aspart with 165 pregnant women with type 1 diabetes treated using human insulin. However. Chlorpropamide and tolbutamide have been shown to cross the placenta. It is not known whether rosiglitazone can cross the placenta. although another four case reports attributed congenital malformations to tolbutamide.36) in the aspart group. [EL = 1++] A small RCT randomised 27 women with gestational diabetes to either insulin aspart or human insulin. the reference guide suggested that insulin is still the treatment of choice. The insulin analogues currently licensed for use in the UK are the rapid-acting analogues lispro. Several case series looked at the incidence of congenital malformations and found that chlorpropamide did not appear to be related to congenital abnormalities in pregnant women. and insulin is normally substituted in women with diabetes.77 [EL = 3] The British National Formulary recommends that metformin.77 [EL = 3] The reference guide reported that no studies had investigated the use of pioglitazone in pregnant women. Prandial increments (mean: breakfast. fetal death and growth retardation.77 [EL = 3] The reference guide stated that case reports had investigated the use of chlorpropamide and tolbutamide in pregnant women with diabetes.80 The trial period extended from the diagnosis of gestational diabetes (18–28 weeks) to 6 weeks postpartum. Sulphonylureas can lead to neonatal hypoglycaemia. although some research is in progress. lunch and dinner) were significantly lower with insulin aspart in the first and third trimester.

The rate of babies with a cranial : thoracic circumference ratio between the 10th and the 25th percentile was significantly higher in the group treated using regular insulin compared with the lispro and control groups. P < 0.84 The review showed no differences between treatments in pregnant women with type 1 diabetes. The 1 hour postprandial blood glucose values were significantly higher in the regular group than in the lispro or control groups. There were no other differences between the three groups in neonatal outcomes. [EL = 1+] Insulin lispro: An in vitro perfusion study of the transfer of insulin lispro across the placenta found no transfer at or below a maternal concentration of 200 microU/ml (corresponding to 26 units of insulin).Diabetes in pregnancy trial period (0.7 kg.7 ± 0. women with gestational diabetes. biochemical hypoglycaemia was significantly more frequent in women who used insulin analogue compared with those who used regular insulin (P < 0. while there were no significant differences between the treatment groups in terms of glycaemic control during the rest of the day.8% versus 6. [EL = 1+] An observational study compared ten pregnant women with type 1 diabetes treated using insulin aspart with ten pregnant women treated using human insulin. [EL = 2+] A pilot study randomised five women with gestational diabetes to receive insulin aspart and five women to receive human insulin. weight. HbA1c values declined during the study period and were similar in both groups. One study that assessed women with gestational diabetes found the total number of hypoglycaemic events did not differ between the groups. The safety profile for both groups was similar. P = 0.8 ± 0. There were no adverse maternal or fetal outcomes in either group. or people with type 2 diabetes.3 ± 0. One study included in the review assessed pregnant women with type 1 diabetes and found the reduction in HbA1c levels to be similar with rapid-acting insulin analogues and regular insulin.17% to –0.9%. There was no perinatal mortality in either group. Biochemical hypoglycaemia (blood glucose less than 3. P < 0. One study involving young women with type 1 diabetes did not show any significant reduction in HbA1c levels between rapid-acting insulin analogues compared with regular insulin. one woman in the regular insulin group developed proliferative retinopathy.5 kg versus 3. insulin dose or mean birthweight.6 ± 1.1 ± 2.0 ± 1. [EL = 1+] An open RCT involving 33 women with type 1 diabetes assessed the effectiveness and safety of preprandial administration of insulin lispro and regular rapid-acting insulin.83 A small placental transfer was observed at maternal concentration of 580 microU/ml (75 units) and 1000 microU/ml (130 units) when maintained for 60 minutes (this duration does not occur in the clinical setting due to the short elimination half-life of insulin lispro).3 ± 1. respectively).12% (95% CI –0. [EL = 1+] 46 .7%. There was no significant difference in birthweight between the insulin aspart and human insulin groups (3. Twenty studies involving women with type 1 diabetes assessed posttreatment HbA1c levels and found weighted mean difference of HbA1c values to be –0. duration of diabetes or gestational age at the time of booking.13). However.0%. [EL = 1++] An RCT that was included in the systematic review provided additional data not reported by the systematic review. Blood glucose was significantly lower (P < 0. HbA1c was lower in the insulin aspart group at booking (7. [EL = 3] A systematic review summarised 42 RCTs that compared rapid-acting insulin analogues (lispro and aspart) with regular insulin in 7933 people (including pregnant women) with type 1 diabetes.4% for human insulin).89 ± 0. type 2 diabetes and gestational diabetes. P < 0. throughout pregnancy and at birth (5.4 versus 5. The study suggests that it is possible to achieve at least as adequate glycaemic control with lispro as with regular insulin therapy in pregnant women with type 1 diabetes.07%) in favour of rapid-acting insulin analogues compared with regular insulin. P < 0.7 versus 5.81 Groups did not differ in terms of age.5 ± 1.3 mmol/litre.5% for the insulin aspart group versus 0.0% versus 8.5% versus 3.0 mmol/litre) was more frequent in the insulin lispro than in the regular insulin group (5.1 ± 0. FBG values (4.05).4 ± 0.01) and pre-lunch blood glucose values (4.86 Blood glucose was determined six times daily and HbA1c every 4 weeks.05).0 mmol/litre.85 In this study 24 women with gestational diabetes treated using insulin lispro were compared with 24 women treated using regular human insulin and 50 women with normal glucose challenge test (GCT).82 There was no difference between the two groups in mean plasma glucose level. Retinopathy progressed in both groups.05).01) were lower with insulin aspart.01) after breakfast in the insulin lispro group.4 ± 2.

7 ± 0. Nonetheless the development of proliferative retinopathy in people with no retinopathy is rare. [EL = 3] Insulin glargine: An observational study of the use of glargine in 22 centres in the UK reported outcomes from 122 babies in 127 pregnancies. The decrease in HbA1c between the first and third trimester was greater in the glargine group (P = 0. Hypoglycaemia requiring assistance occurred in nine (7%) and 16 (12%) had two or more episodes of hypoglycaemia. There is no theoretical basis for an increased risk of retinopathy with insulin lispro apart from facilitating rapid and/or substantial improvement in glycaemic control. HbA1c values declined during the study period and were similar in both groups.5%) women treated using regular insulin. [EL = 1+] Four case series of women with type 1 diabetes exposed to insulin lispro in pregnancy (n = 696) were published since the systematic review. pregnancy complications.89–92 These studies found that insulin lispro improved glycaemic control with no adverse maternal or fetal effects. two occurring in women taking glargine before pregnancy giving a congenital malformation rate of 2. One study reported the development of proliferative retinopathy in three women without background retinopathy at the beginning of pregnancy. The women receiving insulin lispro had significantly lower glucose excursions after a test meal and experienced fewer episodes of hypoglycaemia before breakfast. duration of diabetes and diabetic complications. There was no difference in obstetric or fetal outcomes.5%.88 No other cases of progression of retinopathy in association with insulin lispro have been reported. Background retinopathy developed in one woman. Two of these studies also reported fewer maternal hypoglycaemic episodes and one study reported greater maternal satisfaction with insulin lispro compared with regular insulin.04) and there was a tendency towards 47 . The study suggests that it is possible to achieve at least as adequate glycaemic control with lispro as with regular insulin therapy in pregnant women with type 1 diabetes. progressed in three women and laser photocoagulation was required in seven women. one woman in the regular insulin group developed proliferative retinopathy. [EL = 2++] An open RCT involving 33 women with type 1 diabetes assessed the effectiveness and safety of preprandial administration of insulin lispro and regular rapid-acting insulin. parity.2% at booking to 6. HbA1c fell from 8. although it has been observed in association with rapid and substantial improvement in glycaemic control. Retinopathy progressed in both groups. The review included an RCT of 42 women with insulin-requiring gestational diabetes randomly allocated to either insulin lispro or regular insulin.1%) with pre-existing type 1 or type 2 diabetes who had been exposed to insulin lispro during embryogenesis. while there were no significant differences between the treatment groups in terms of glycaemic control during the rest of the day.1% during the third trimester. ventricular septal defect and transposition of the great arteries (TGA)).01) after breakfast in the lispro group. In comparison there were 14 congenital malformations in 133 (10. respectively).94 Groups were matched for age.Preconception care An earlier systematic review87 was found on the use of insulin lispro in pregnancy. [EL = 3] An observational study compared maternal and perinatal outcomes in 47 women with type 1 diabetes taking glargine before pregnancy to 50 women using long-acting insulin (protaphan).5% versus 3. There were three congenital malformations (positional talipes. There was no perinatal mortality in either group. All 122 babies were liveborn. The aim of the study was to investigate whether insulin lispro crosses the placenta.93 One hundred and fifteen women had type 1 diabetes. They also experienced fewer hyperglycaemic episodes. Biochemical hypoglycaemia (blood glucose less than 3.0 mmol/litre) was more frequent in the insulin lispro group than in the regular insulin group (5. There were seven congenital malformations in 170 women (4. [EL = 1+] The review also included 12 observational studies involving 303 women (294 with type 1 diabetes. Blood glucose was significantly lower (P < 0. five had type 2 diabetes and seven had gestational diabetes. No insulin lispro was detected in umbilical cord blood and there was no difference between the two groups in insulin antibodies. neonatal birthweight. There were seven (6%) early miscarriages.9%. nine with type 2 diabetes). shoulder dystocia or respiratory distress).1 ± 0. However. Three observational studies reported improved glycaemic control in pregnant women using insulin lispro compared with pregnant women using regular insulin. other risk factors for progression of retinopathy were present.86 Blood glucose was determined six times daily and HbA1c every 4 weeks. There was no difference in outcomes (gestational age at birth.

87. including the risk of hypoglycaemia. glargine or detemir in pregnancy. Until sufficiently powered RCTs of insulin lispro in pregnancy have confirmed its safety and effectiveness. although a meta-analysis of observational studies found no increased risk of congenital malformations. during the period of organogenesis. Evidence statement A laboratory study using validated methodology found glibenclamide did not cross the placenta. detemir and glargine) during pregnancy should be avoided until more data are available on their safety. The use of other rapid.95 There was no significant difference between the birthweight or centile birthweight of babies born to the women treated using insulin glargine and those born to the control group treated using intermediate-acting human insulin. i.5% (12/32) in the insulin glargine group and 40. a large number of observational studies of its use during pregnancy have found no teratogenic effect. From evidence to recommendations Oral hypoglycaemic agents have a number of potential advantages over insulin for the treatment of type 2 diabetes during pregnancy and for the treatment of gestational diabetes. the risks and benefits of insulin lispro. however. four women with gestational diabetes) and two case reports (both in women with type 1 diabetes) found no adverse maternal or fetal outcomes associated with the use of glargine in pregnancy. There was no significant difference in neonatal morbidity between the groups. should be weighed on an individual basis. P = 0. [EL = 2+] A matched case–control study involving 64 pregnant women with diabetes investigated the association between insulin glargine use during pregnancy and incidence of fetal macrosomia or adverse neonatal outcomes. promoting the principle of informed choice. They are more convenient and less expensive than insulin and may improve the long-term prognosis of women with type 2 diabetes and gestational diabetes. No high-quality studies have considered the use of oral hypoglycaemic agents in the first 7 weeks of pregnancy. promoting the principle of informed choice. should be weighed on an individual basis.07).6% (13/32) in the control group. Until this evidence is available the risks and benefits of metformin during pregnancy. A large number of studies have shown no indication that insulin lispro is teratogenic. Metformin crosses the placenta.e.96–101 [EL = 3] Existing guidance The NSF for diabetes20 recommends that women with type 2 diabetes who require treatment with oral hypoglycaemic agents to achieve good blood glucose control and who are planning to become pregnant or are already pregnant should be transferred to insulin therapy because of the theoretical risk associated with these drugs crossing the placenta. including the risk of hypoglycaemia. There have been no clinical trials of glulisine. Incidence of fetal macrosomia was 37. An RCT has found glibenclamide to be an effective alternative to insulin for the treatment of gestational diabetes. 48 . For some women insulin analogues may offer benefits over isophane insulin (also known as NPH insulin) in terms of flexibility and improved glycaemic control.Diabetes in pregnancy fewer hypoglycaemic episodes (11/47 in glargine group versus 21/50 in protaphan group. [EL = 2+] Four case series (26 women with type 1 diabetes. The evidence suggests insulin glargine treatment during pregnancy does not appear to be associated with increased fetal macrosomia or neonatal morbidity. These studies are needed before they can be recommended for use in early pregnancy. The safety record of glibenclamide and metformin is sufficient for well-designed RCTs to be conducted on its use in early pregnancy. RCTs and observational studies have shown that insulin aspart is effective for managing diabetes in pregnancy without increasing the risk of hypoglycaemia.and long-acting insulin analogues (glulisine.

Preconception care Recommendations for the safety of medications for diabetes before and during pregnancy Women with diabetes may be advised to use metformin as an adjunct or alternative to insulin in the preconception period and during pregnancy. There were two LGA babies without complications. Mean serum creatinine and uric acid levels increased significantly in the first and third trimesters. Proteinuria was 1292 ± 656 mg/day before treatment. All women maintained blood glucose levels near to normal throughout pregnancy. Creatinine clearance and potassium remained stable. ACE inhibitors are used to treat hypertension and slow the progression of nephropathy. which resolved immediately following birth. 3. There were 11/24 cases (46%) of superimposed  Metformin is used in UK clinical practice in the management of diabetes in pregnancy and lactation. At conception it was 273 ± 146 mg/ day (P < 0. 202 ± 141 mg/day at conception (P = 0. Women attempted conception when proteinuria was less than 500 mg/day and euglycaemia was achieved. Before ACE inhibitor treatment proteinuria was 1633 ± 666 mg/day. There were no research recommendations relating to which medications for diabetes are safe for use during pregnancy and which should be discontinued. 650 ± 502 mg/day in the first and second trimesters. diabetes should not be treated with metformin but insulin should be used to maintain blood glucose levels. In two women (25%). The SPC advises that when a patient plans to become pregnant and during pregnancy. Therefore isophane insulin (also known as NPH insulin) remains the first choice for long-acting insulin during pregnancy. [EL = 3] In a case series of 24 women with diabetic nephropathy and normal to mild renal impairment104 all women received ACE inhibitors for at least 6 months prior to conception and maintained strict glycaemic control from at least 3 months prior to conception. Three cohort studies and two case series were identified that considered the maternal. All other oral hypoglycaemic agents should be discontinued before pregnancy and insulin substituted. There was no significant change in any other renal function test (creatinine clearance test. Healthcare professionals should be aware that data from clinical trials and other sources do not suggest that the rapid-acting insulin analogues (aspart and lispro) adversely affect the pregnancy or the health of the fetus or newborn baby. There were no other adverse maternal or neonatal outcomes. Informed consent on the use of metformin in these situations should be obtained and documented. Proteinuria increased gradually over each trimester to 593 ± 515.001). At conception ACE inhibitors were discontinued. potassium). Women remained euglycaemic throughout pregnancy. There is strong evidence for its effectiveness and safety.2 between trimesters). Three months after birth proteinuria was 619 ± 411 mg/day. uric acid. This evidence is not currently reflected in the SPC (July 2008). 1012 ± 1206 mg/day in the third trimester and 590 ± 410 mg/day 6–8 weeks after birth. 49 . There were three cases of pre-eclamptic toxaemia (PET) just prior to birth. fetal and neonatal effects of first-trimester exposure to ACE inhibitors. 783 ± 813 and 1000 ± 1185 mg/day. when the likely benefits from improved glycaemic control outweigh the potential for harm. serum creatinine.7 The safety of medications for diabetic complications before and during pregnancy Description of the evidence Angiotensin-converting enzyme inhibitors In people with diabetes.102 Two case series were identified that reported on the use of ACE inhibitors in women with diabetes and nephropathy who were planning a pregnancy and who discontinued treatment on confirmation of pregnancy.0001). Angiotensin-converting enzyme inhibitors before pregnancy: In eight women with diabetic nephropathy103 ACE inhibitors were used for a minimum of 6 months prior to conception. proteinuria exceeded 1000 mg/day during pregnancy. respectively (P = 0. Women with insulin-treated diabetes who are planning to become pregnant should be informed that there is insufficient evidence about the use of long-acting insulin analogues during pregnancy.

77 [EL = 3] The British National Formulary recommends that ARBs should be avoided during pregnancy as they may adversely affect fetal and neonatal blood pressure control and renal function.72 to 4.1. There were two cases of IUGR.27) as compared with babies who had no exposure to antihypertensive medications. One baby was admitted to intensive care because of prematurity. [EL = 2++] A report of post-marketing surveillance for ACE inhibitors108 reported outcomes of pregnancy in 66 women who self-enrolled to the registry after first-trimester-only exposure.e. There were 48 live births and 15 miscarriages (23%).5%. moexipril. which included the date when the prescription was filled and the number of days for which the medicine was supplied. Babies with firsttrimester-only exposure to ACE inhibitors had an increased risk of major congenital malformations (RR 2. Other known risk factors were present in the three cases of IUGR (i. one of which ended in an intrauterine death. There were no major malformations. There was no information about captopril.66. There was one intrauterine death at 24 weeks of gestation secondary to early severe IUGR. multiple gestation or hypertension).106 No congenital abnormalities or neonatal renal dysfunction were reported. were excluded. perindopril. Babies of women with diabetes. cilazapril. they may also cause skull defects and oligohydramnios. 50 .75). 95% CI 0. [EL = 3] Angiotensin-converting enzyme inhibitors in the first trimester: A cohort study105 included 29 507 babies. Among the 48 live births there were three cases of IUGR. In contrast. 95% CI 1. imidapril or ramipril.25 to 1. Angiotensin-II receptor blockers A reference guide to drugs in pregnancy and lactation reported that data from studies involving pregnant women suggest a risk to the fetus if ARBs are used in the second or third trimesters of pregnancy.107 There were no stillbirths or congenital malformations among the 21 babies. or who were exposed to ACE inhibitors after the first trimester. Even in the six pregnancies in which ACE inhibitors were continued beyond 12 weeks of gestation (including one where therapy was continued until 25 weeks of gestation) there were no congenital abnormalities. Another child had a patent ductus arteriosus that required surgical ligation at 18 months. [EL = 2++] A third cohort study assessed the effect of prescription of ACE inhibitors at 5–15 weeks of gestation in 21 pregnant women in terms of adverse pregnancy outcomes. quinapril and trandolapril are used in the second or third trimesters of pregnancy. Major congenital malformations were identified from medical records. fosinopril. This was attributed to severe disease in the mother rather than drug effect. [EL = 3] A reference guide to drugs in pregnancy and lactation reported that data from studies involving pregnant women suggest a risk to the fetus if the ACE inhibitors enalapril.Diabetes in pregnancy pre-eclampsia.77 [EL = 3] The British National Formulary recommends that ACE inhibitors should be avoided during pregnancy as they may adversely affect fetal and neonatal blood pressure control and renal function and they may cause skull defects and oligohydramnios. Maternal use of prescribed medications was determined from pharmacy files. [EL = 3] A prospective case series109 reported outcomes for eight women treated using ACE inhibitors in the first trimester. exposure to other antihypertensive medications during only the first trimester did not confer an increased risk (RR 0.78 There is information on the use of ACE inhibitors during breastfeeding in Section 8. The rate of caesarean section was 62. [EL = 2+] A second cohort study which assessed ACE inhibitor use before pregnancy and in early pregnancy assessed adverse pregnancy outcomes in 18 women and 19 pregnancies. 4/24 cases (17%) of preterm birth and 5/24 cases (21%) of intrauterine growth restriction (IUGR). 209 of whom had been exposed to ACE inhibitors in the first trimester alone and 202 of whom had been exposed to other antihypertensive medication in the first trimester. No deterioration was observed in any of the women at 2 year follow-up.71. or who were exposed to any other potential teratogens. Two babies had cerebral palsy due to birth trauma secondary to LGA.78 There is information on the use of ARBs during breastfeeding in Section 8.1. lisinopril.

rosuvastatin and simvastatin.78 There is information on the use of statins during breastfeeding in Section 8. The decrease in blood pressure was smooth and controlled and was associated with an insignificant increase in heart rate. There were limited data for the use of nifedipine in pregnant women. and the reference guide suggested that it presents a low risk to the fetus.27). but the risk to the fetus should be balanced against the risk of uncontrolled maternal hypertension. as evidenced by cardiotocographic monitoring. The case series evaluated 20 cases of malformation in 54 cases of statin exposure reported to the US Food and Drug Administration between 1987 and 2001. There were limited data for the use of felodipine. There were limited data for the use of isradipine in pregnant women and the reference guide suggested that it presents a low risk to the fetus. pravastatin and simvastatin in pregnant women. This study suggests that calcium-channel blockers do not represent a major teratogenic risk. There was no review of lacidipine or lercanidipine. [EL = 2++] Another cohort study investigated the effect of verapamil infused intravenously after plasma volume expansion with dextran-70 in nine women with severe gestational proteinuric hypertension. P = 0.Preconception care Statins Statins (HMG CoA reductase inhibitors) are used to reduce elevated levels of cholesterol. attributed to maternal disease by stepwise regression. There were no adverse fetal effects. nonteratogenic controls 0%. A reference guide to drugs in pregnancy and lactation reported that atorvastatin. Pregnancy outcome was compared with that of a control group matched for maternal age and smoking habits. P = 0.111 The haemodynamic response in the women and adverse fetal effects were monitored. The reference guide suggested that verapamil is compatible with pregnancy. was the most important factor responsible for the observed decrease in birthweight (mean −334 g versus nonteratogenic controls. [EL = 3] The British National Formulary recommends that statins should be avoided during pregnancy as congenital malformations have been reported and decreased synthesis of cholesterol may affect fetal development.110 Six teratogen information services prospectively collected and followed up 78 women with firsttrimester exposure to calcium-channel blockers.08).1. P = 0. Felodipine may inhibit 51 . The defects reported were attributable to maternal diabetes or co-ingestion of teratogens.78 Amlodipine and nimodipine have no information available about possible harms. [EL = 2+] A reference guide to drugs in pregnancy and lactation reported that there were limited data for the use of diltiazem in pregnant women. and suggested that it presents a high risk to the fetus. There was no increase in major malformation rates (calcium-channel blockers 3.0% (2/66).77 [EL = 3] The British National Formulary suggests that the calcium-channel blocker verapamil may reduce uterine blood flow leading to fetal hypoxia and that it may inhibit labour. and the reference guide suggested that they present a risk to the fetus. Statins currently licensed for use in the UK include atorvastatin. the manufacturers advise pregnant women to avoid them. nonteratogenic controls 9%. Isradipine. There were no studies investigating the use of amlodipine or nisoldipine in pregnant women. The malformations included five major defects of the central nervous system (including two cases of holoprosencephaly) and five unilateral limb deficiencies. the manufacturer advises women to avoid it in the first trimester of pregnancy unless absolutely necessary. The increase in preterm birth (calcium-channel blockers 28%.77 [EL = 3] The reference guide reported that a small number of case reports and surveillance studies and a case series had investigated the use of atorvastatin. pravastatin and simvastatin are contraindicated in pregnancy and lactation. fluvastatin. The apparent effectiveness of verapamil in this study justifies further investigation. Verapamil produced a statistically significant reduction in mean arterial pressure and systemic vascular resistance without adversely affecting cardiac output. nicardipine or nimodipine in pregnant women. Calcium-channel blockers A cohort study examined the potential teratogenicity of calcium-channel blockers. but the risk to the fetus should be balanced against the risk of uncontrolled maternal hypertension. and the reference guide suggested that these present a moderate risk to the fetus. fluvastatin. There was no review for rosuvastatin. the manufacturers advise pregnant women to avoid them. fluvastatin. pravastatin.003). nifedipine and nicardipine may inhibit labour.

No clinical evidence was identified in relation to the safety of obesity drugs in pregnancy. The British National Formulary recommends that orlistat should be used with caution in pregnancy and that other drugs for the treatment of obesity should be avoided. There is information on the use of calcium-channel blockers during breastfeeding in Section 8.77 statins should be avoided during pregnancy. Lercanidipine and diltiazem have no information available. Obesity drugs A reference guide to drugs in pregnancy and lactation reported that there were no data for the use of orlistat (a lipase inhibitor) or sibutramine (a centrally acting appetite suppressant) in pregnant women. Renal function did not appear to deteriorate during pregnancy. Lacidipine may inhibit labour. The British National Formulary recommends that ARBs and statins should be avoided during pregnancy. However a large. the risk of progression to macroalbuminuria in the preconception period is thought to be small while the risk of pre-eclampsia is greatly increased. There was no evidence of teratogenic effect and evidence of moderate-to-good pregnancy outcomes associated with the use of ACE inhibitors. the manufacturer advises pregnant women to avoid them. From evidence to recommendations ACE inhibitors and ARBs (angiotensin-II receptor antagonists) should be avoided throughout pregnancy because of the possible risk of congenital malformations. No clinical evidence was identified in relation to the safety of ARBs or statins in pregnancy. and it suggested that they present a low risk to the fetus. Two small case series reporting on the use of ACE inhibitors in the first trimester of pregnancy did not detect a teratogenic effect. Nisoldipine should be avoided by pregnant women. but no large-scale trials of their effectiveness and safety in pregnancy were identified. As cholesterol and products synthesised during pregnancy are important to fetal development and as there is no apparent harm from interrupting cholesterol-lowering therapy during pregnancy.1.77 [EL = 3] The British National Formulary reports that the manufacturers of orlistat advise that it should be used with caution during pregnancy. the benefits of continuing with ACE inhibitors until discontinuation of contraception for the purposes of protecting renal function should be considered. The manufacturers of sibutramine and rimonabant recommend that they should be avoided in pregnancy. There was no review for rimonabant. For women with microalbuminuria.78 There is information on the use of obesity drugs during breastfeeding in Section 8.1. However. Calcium-channel blockers should be avoided throughout pregnancy because of the risk of disruption to labour and fetal hypoxia. the risk to the fetus should be balanced against the risk of uncontrolled maternal hypertension in deciding whether to discontinue nifedipene.Diabetes in pregnancy labour and pregnant women are advised to avoid it. the manufacturer advises pregnant women to avoid it. well-controlled cohort study that linked pregnancy outcomes to prescription records found that babies with first-trimester-only exposure to ACE inhibitors had an increased risk of major congenital malformations compared with babies who had no exposure to antihypertensive medications. [EL = 3] 52 . Two small cohort studies suggest that calcium-channel blockers do not have a teratogenic effect. Evidence statement Two small case series reported on the use of ACE inhibitors until conception in women with diabetic nephropathy. The British National Formulary recommends that they should be avoided in pregnancy. However. another centrally acting appetite suppressant.

access to health care. and in which the woman stated that she attempted to achieve optimal blood glucose control before becoming pregnant.2 [EL = 3] A 2002 audit of units expected to provide maternity care for women with diabetes in England.Preconception care Recommendations for the safety of medications for diabetic complications before and during pregnancy Angiotensin-converting enzyme inhibitors and angiotensin-II receptor antagonists should be discontinued before conception or as soon as pregnancy is confirmed.8 Removing barriers to the uptake of preconception care and when to offer information Preconception care is aimed at reducing congenital abnormalities by improving maternal glycaemic control before conception and during the first 7 weeks of pregnancy. a self-administered questionnaire (for information on demographics. contraceptive behaviour and including the Marital Satisfaction Scale and the Health Locus of Control scale). [EL = 2+] 53 .001). 3. A CEMACH audit of pregnancy in women with pre-existing diabetes in England. There were no research recommendations relating to which medications for diabetic complications are safe for use during pregnancy and which should be discontinued.112 Nonparticipants included women excluded from the study due to adverse pregnancy outcome. and interviews (for responses on a range of topics potentially related to pregnancy planning behaviour). Statins should be discontinued before pregnancy or as soon as pregnancy is confirmed. Description of the evidence A CEMACH audit from England. in 49% there was advice within the general diabetes clinic only. ‘Preconception care’. 13% at a preconception clinic and 4% from the GP (the service providing the counselling was not known for 15% of women). in which contraception was stopped or avoided for the purposes of becoming pregnant. or advice by an obstetrician within the obstetric services only. dietary advice and glucose monitoring • preparation for pregnancy. the remaining units had other arrangements. Wales and Northern Ireland (n = 3808) found that only 35% of women received preconception counselling. A planned pregnancy was defined as a pregnancy that was desired before conception. Alternative antihypertensive agents suitable for use during pregnancy should be substituted. All other pregnancies were defined as unplanned. Women with type 2 diabetes and women from a minority ethnic group were significantly less likely to have had a measure of long-term glycaemic control (P < 0. The results were as follows. 17% of units had separate preconception clinics. including supplementation with folic acid. By this definition 41% (35/85) of pregnancies were planned.32 [EL = 3] A population-based study recruited 85/122 (70%) women with diabetes who gave birth at 15 participating hospitals over 12 consecutive months. 68% received counselling at the adult diabetes clinic. Of these. and assessment and treatment (if necessary) of diabetic complications. ‘preconception counselling’ and ‘preconception clinic’ are often used interchangeably but most definitions include the following components: • information and education for the woman and her partner • support in improving glucose control before pregnancy through intensified insulin regimens. advice within the general diabetes clinic and by an obstetrician within obstetric services. Folic acid was taken by 39% of women and 37% were reported as having a measure of long-term glycaemic control in the 6 months prior to pregnancy. Wales and Northern Ireland and four studies undertaken in the USA were found that looked at the factors influencing uptake of preconception care in women with diabetes. medication review. Study methods included: a review of medical records (for HbA1c on entry to preconception care). Wales and Northern Ireland reported that there was no formal arrangement for preconception counselling in 16% of units.

49% (17/35) had a previous pregnancy with diabetes. In the women with planned pregnancies. Knowledge Nearly all women with planned pregnancies (33/35. 80% (28/35) of women believed that their partners were well informed about diabetes and pregnancy issues before the pregnancy. 54 . 14% (7/50) of women with unplanned pregnancies received reassuring advice. to have attended for diabetes care in the 6 months before conception (P = 0. 94%) and 68% (34/50) of women with unplanned pregnancies knew they should be in good diabetes control before pregnancy. Eight women with unplanned pregnancies could not recall hearing any information before pregnancy about diabetes and pregnancy. 56% (28/50) had a previous pregnancy with diabetes. However. women with planned pregnancies were more likely to understand the specific association between high blood glucose levels and birth defects (83% versus 30%). P = 0. to have private health insurance (P < 0. Among women with unplanned pregnancies.8.0001). Seventy percent (35/50) of women with unplanned pregnancies said they were very happy when they found out they were pregnant. [EL = 2+] A multicentre case–control study113 compared women with pre-existing diabetes making their first preconception visit (n = 57. were more likely to be white (P = 0. Contraceptive behaviour and desire for motherhood Among the 50 women with unplanned pregnancies. Relationship with partner Women with planned pregnancies were more satisfied with their relationship with their partner than were women with unplanned pregnancies. 75% (26/35) felt they had received reassuring and encouraging advice from their providers before pregnancy.28. There was no association between pregnancy planning and internal locus of control (the belief that health outcomes are largely under one’s own control) or chance locus of control (the belief that health is due to luck or fate). 18 type 2 diabetes).007).004). 53 type 1 diabetes.003) and were less likely to smoke (P < 0. Personality Women with unplanned pregnancies were significantly more likely to have an external locus of control (i. 70% (35) used contraception less than half the time. In women with unplanned pregnancies. Women with planned pregnancies had significantly higher income (P < 0.Diabetes in pregnancy Background characteristics The average SD above the laboratory mean for glycohaemoglobin at the first antenatal visit was significantly lower in planned than unplanned pregnancies (3.1 versus 5. 79 type 1 diabetes. four type 2 diabetes) with those making their first antenatal visit without having received preconception care (n = 97. to attribute their health outcomes to the control of powerful others) than were women with planned pregnancies (OR 2. 38% (19/50) recalled that pregnancy was discouraged. In the planned pregnancy group.004).001).05). Advice from healthcare provider Among women with planned pregnancies. Women who described a positive relationship ‘felt it was important that their doctor understood the difficulty of living with diabetes and did not judge them on their blood glucose control’.0002). Many couples had attended appointments together and almost all women expressed a feeling of being supported. Relationship with healthcare provider A positive relationship with their healthcare provider was described by 75% (25/35) of women with planned pregnancies compared with 28% (14/50) of women with unplanned pregnancies.001).e.81. In contrast. After logistic regression the following variables were associated with seeking preconception care: education (OR 4. more education (P = 0. In the antenatal group only 24% of pregnancies were planned. (There were five women who thought that diabetes made it more difficult to conceive). P = 0. Logistic regression found a significant association between Marital Adjustment Scale score and pregnancy planning (OR 3. In contrast 16% (8/50) of women with unplanned pregnancies felt their partners were informed about diabetes and pregnancy before the pregnancy and most ‘felt that their partners did not understand the enormity of effort required to achieve good diabetes control’. P < 0.86.

social support that involved practical. women applied the term ‘planned’ only when four criteria were met. P < 0.03). When asked to apply the terms.05) and positive attitudes towards contraception (P < 0. The study also found that women with planned pregnancies were significantly more likely to have a partnership rated ‘high’ in overall quality (χ² = 7. P < 0.01). solidification of an unstable relationship. 48% (16) of which were unplanned. Consistent use of contraception was significantly associated with higher levels of social support for contraception (P < 0. values between 6% and 8% were considered ‘adequate’ and values greater than 8% were considered ‘sub-optimal’. In the prevention of unplanned pregnancy stage 82% of women with diabetes and 88% of women without diabetes used contraception. Overall 29 women stated that they had planned their pregnancies but only 12 (22%) saw a physician before conception to modify insulin intake or glycaemic control. or (4) treatment continuation with pregnancy.41.e. beliefs and behaviours regarding diabetes in pregnancy. The term ‘unplanned’ covered a wide variety of circumstances.05).02) and adherence to diabetes regimen (OR 3. Logistic regression revealed that not being advised to achieve target glucose or HbA1c values (questionnaire response) was associated with entering pregnancy with poor glycaemic control (P = 0. There was no significant association with knowledge of contraception. (3) commencement of intensive metabolic control. knowledge of diabetes. The study used annual questionnaires and interviews to assess knowledge. [EL = 2+] An interview study with 47 women without diabetes117 found that. Intending to become pregnant and stopping contraception were not sufficient.05). for example by alleviating previous problems. Women with unplanned pregnancies were significantly more likely to be rated as having a ‘high’ chance of secondary gain than women with planned pregnancies (χ² = 29. quality of relationships. All women were given a questionnaire on preconception education and control of glycaemia. personality and social support. Women with diabetes were more likely to use condoms (P = 0. P < 0. P < 0. [EL = 2+] A 5 year longitudinal study involved 66 women with type 1 diabetes and 207 women without diabetes. women did not use the terms ‘planned’ or ‘unplanned’ spontaneously.Preconception care P < 0. self-esteem and the ‘secondary gain’ inherent in becoming pregnant were examined. Women in the preconception care group were significantly more likely to perceive that preconception care conferred benefits to the woman and child and to report instrumental social support (i. An interview study using a semi-structured questionnaire included 88 pregnant women and 40 non-pregnant women.01).25.115 HbA1c levels were determined either before conception or during the first trimester. Examples of secondary gain included autonomy from parents. when discussing the circumstances of their pregnancies. Themes that emerged during interviews included the belief that oral contraceptives were unsafe for women with diabetes and that diabetes made it difficult to conceive.05). Sixty percent (33/55) of women had sub-optimal control while only 11% (6) had values in the normal range. although two studies were found in women without diabetes. attitudes.10. Secondary gain was used to refer to advantage or benefit that may arise from becoming pregnant.116 Among the pregnant women 45 had a planned pregnancy and 43 had an unplanned pregnancy. attitudes. tangible aid offered by another individual). Women were classified as being in one of four stages of the maternal diabetes lifecycle: (1) prevention of unplanned pregnancies. Women with prior poor outcome of pregnancy were significantly more likely to enter pregnancy with poor glycaemic control (P = 0.25.114 All women with diabetes were at least 1 year post diagnosis. encouraged by provider to receive preconception care (OR 3.0001). However these items were not significant when entered into a logistic regression model. Agreeing to the pregnancy with their partner and reaching the right time in terms of lifestyle or life stage were also necessary. an excuse to leave a boring job and/or begin a new and important phase in their lives. [EL = 2+] A cross-sectional study surveyed 55 women with pre-existing diabetes attending a preconception clinic. All women in the study completed a validated questionnaire on knowledge. [EL = 2+] No comparable studies were found for the UK. living with partner (OR 11. In the 5 years of the study there were 23 pregnancies in women with diabetes. The authors concluded that women with unplanned pregnancy may in fact fall into a group of ‘semi-planned’ pregnancy ‘who are indifferent or at worst ambivalent about the idea of pregnancy’. Values in the normal range (4–6%) were considered ‘optimal’.01). 78% (17) of which were unplanned. visited a diabetes clinic in the last year (OR 8. (2) reproductive decision making. difficulties.39. P < 0. Life events. In women without diabetes there were 33 pregnancies. locus of control or self-esteem.02). [EL = 2+] 55 .02). P < 0.03.

Evidence statement The following barriers to preconception care were identified. attitudes.Diabetes in pregnancy Young women A case–control study118 examined knowledge. The study results showed that having diabetes did not appear to significantly decrease the risk-taking behaviour of the young women. [EL = 3] Current practice A CEMACH survey of maternity units in England. motivation and support with regard to diabetes self-management and pregnancy planning. Wales and Northern Ireland assessed the quality of maternity service provision against standards set out in the NSF for diabetes. For example.2. attitudes. an unsupportive partner. Knowledge Common misconceptions are that diabetes decreases fertility and that oral contraceptives have a detrimental effect on diabetes. in particular. sexual activity and birth control in young women with diabetes as well in those without diabetes. of the term ‘preconception counselling’ and its role in preventing such complications. Pregnancy planning Most pregnancies in women with diabetes are unplanned. The study showed that young women with diabetes lacked awareness of pregnancy-related complications associated with diabetes. Knowledge of the effects of the benefits of good glycaemic control prior to pregnancy does not guarantee pregnancy planning. Social support A lack of social support. one study found that the majority of women with unplanned pregnancy were not using contraception and said they were very happy when they found out they were pregnant. Sociocultural factors Women with unplanned pregnancies are more likely to be from a lower socio-economic group or a minority ethnic group. is associated with unplanned pregnancy. [EL = 2+] Further information about current practice in terms of preconception care and the importance of planning pregnancy is presented in Section 3. This reflects complex social and cultural differences in knowledge. One study found that 16% of women with an unplanned pregnancy could not recall hearing any information before pregnancy about diabetes and pregnancy. Relationship with healthcare professionals Feeling judged and being discouraged from becoming pregnant is associated with unplanned pregnancy. That some ‘unplanned’ pregnancies may not have been unexpected or unwanted. and of the importance for women with diabetes to use a highly effective method of contraception for preventing unplanned pregnancy. in one study more than half of unplanned pregnancies were in white women with tertiary education. One study found nearly half of the women with unplanned pregnancy had a previous pregnancy with diabetes. 56 .32 It found that in 2002 only 17% of maternity units provided a preconception clinic. this proportion had remained largely unchanged from 1994 (16%). intentions and behaviours regarding diabetes and reproductive issues. The evidence suggests that pregnancy planning should be considered a continuous rather than a dichotomous variable – only a minority of unplanned pregnancies are due to contraceptive failure. Nonetheless. reflects ‘conflicting or ambivalent emotions’ towards motherhood112 or an underlying desire to bring about change in other areas of life. [EL = 2−] A descriptive qualitative study119 examined whether young women with type 1 diabetes were aware and concerned about pregnancy-related complications and how to prevent them.

they lacked awareness of the importance of planning pregnancy. To overcome barriers to preconception care information about the importance of planning pregnancy and achieving good glycaemic control in the periconceptional period should be reinforced at each contact between women with diabetes who are of childbearing age and their healthcare professionals. Having diabetes did not reduce risk-taking behaviour in young women and. These showed that the reduction in major congenital malformations needed for preconception care and advice to be considered costeffective was much lower than reported in a meta-analysis of cohort studies of preconception care 57 . Preconception care for women with diabetes should be given in a supportive environment and the woman’s partner or other family member should be encouraged to attend. A retrospective cohort study120 examined the effect of an intensive diabetes management programme during pregnancy on women’s long-term self-management behaviours and glycaemic control. frequency of insulin injections and frequency and complexity of insulin dose adjustment from entry to the programme to birth of the baby. The methods and results from the health economic modelling are summarised here. The GDG approached the analysis by considering the cost-effectiveness of preconception care and advice for women with diabetes. from adolescence. the robustness of baseline results were assessed using threshold analyses. Recommendations for removing barriers to the uptake of preconception care and when to offer information Women with diabetes should be informed about the benefits of preconception glycaemic control at each contact with healthcare professionals. 3. including their diabetes care team. There was also a significant improvement in HbA1c from entry to birth of the baby. further details are provided in Appendix C.9 Cost-effectiveness of self-management programmes The effectiveness of self-management of diabetes in pregnancy was identified by the GDG as a priority for health economic analysis. Due to data limitations and uncertainty about the effectiveness of different forms of preconception care and advice. [EL = 2−] An economic model constructed for the purposes of this guideline suggested that some form of preconception care and advice was likely to be cost-effective. as for other women with diabetes. Young women A case–control study and a qualitative study have investigated attitudes to pregnancy in young women with diabetes. From evidence to recommendations Evidence shows that women with diabetes (including young women) lack awareness of the importance of planning pregnancy and the role and purpose of preconception care. The intentions of women with diabetes regarding pregnancy and contraceptive use should be documented at each contact with their diabetes care team from adolescence. The study showed that there was a significant improvement in all diabetes self-management behaviours including frequency of self-monitoring of blood glucose. including the diabetes care team. In particular. a service that is provided only after women indicate that they are planning to become pregnant will have limited uptake.Preconception care Appropriateness and availability of services Only a minority of maternity units provide a preconception clinic and services currently provided may not be effective in addressing the barriers to uptake of preconception care identified above. and the woman’s intentions regarding pregnancy in the immediate future or longer term should be discussed. There were no research recommendations relating to the barriers to uptake of preconception care and the information that should be offered to all women of childbearing age with diabetes and/or women with diabetes who are planning a pregnancy.

org. available from www.Diabetes in pregnancy and advice. Diabetes Education and Self Management for Ongoing and Newly Diagnosed type 2 diabetes (DESMOND) and X-PERT for people with type 2 diabetes. Women with diabetes who are planning to become pregnant should be offered preconception care and advice before discontinuing contraception. From evidence to recommendations NICE recommends that structured education programmes are made available to people with type 1 and type 2 diabetes. are likely to be even greater for women planning a pregnancy because good glycaemic control improves pregnancy outcomes. the different methods of delivering preconception care and advice vary enormously in terms of their resource use and ideally the incremental costeffectiveness ratios (ICERs) of these alternative methods should be compared.121 However. available from www. rather than services provided by trained counsellors.62 There is some concern that confounding may explain at least some of the observed effect and that the potential benefits in term of improved glycaemic control may be lessened in settings where structured education programmes are offered more generally to women with diabetes.org.18 The benefits of such programmes. Future research should seek to establish the clinical and costeffectiveness of different models of preconception care and advice for women with pre-existing  ‘Type 2 diabetes: the management of type 2 diabetes’ (NICE clinical guideline 66). the GDG did not recommend a specific form of preconception care and advice. However. 58 . which include DAFNE69 for type 1 diabetes.uk/TA060. A meta-analysis of preconception care and advice found that it conferred a statistically significant reduction in the rate of major congenital malformations in offspring born to women with diabetes. The GDG considers. Due to limitations in the clinical evidence available to inform the health economic modelling it was not possible to establish the optimal form of preconception care and advice for this group of women.nice. Therefore.nice. updates the information on type 2 diabetes in this technology appraisal. such an analysis is not really possible using existing data and their inherent limitations.121 [EL = 2++] and that cost-effectiveness of preconception care and advice relative to no preconception care and advice was not very sensitive to the costs of preconception care and advice or the ‘downstream’ costs associated with major congenital malformations. Preconception care and advice can take many different forms and some of this heterogeneity was reflected in the meta-analysis. the economic model demonstrated that a much smaller effect size than reported in the metaanalysis would still be considered cost-effective and the GDG considered that preconception care and advice would meet the threshold for cost-effectiveness. Recommendations for self-management programmes Women with diabetes who are planning to become pregnant should be offered a structured education programme as soon as possible if they have not already attended one (see ‘Guidance on the use of patient-education models for diabetes’ [NICE technology appraisal guidance 60]. that it is particularly important that women with diabetes do not forego the benefits of such programmes. The GDG has used the phrase ‘preconception care and advice’ (rather than ‘preconception counselling’) in the recommendations to emphasise that this is care and advice provided by members of the multidisciplinary diabetes and antenatal care team (see Section 5. therefore.uk/CG066.8) during the preconception period. However. Research recommendations for self-management programmes What is the most clinically and cost-effective form of preconception care and advice for women with diabetes? Why this is important Preconception care and advice for women with pre-existing diabetes is recommended because a health economic analysis has demonstrated cost-effectiveness of attendance at a preconception clinic.

referral to a nephrologist should be considered before discontinuing contraception. There were no research recommendations in relation to renal assessment in the preconception period. or the estimated glomerular filtration rate (eGFR) is less than 45 ml/minute/1.5. 3. The research should also seek to establish whether women with type 1 and type 2 diabetes have different needs in terms of preconception care and advice. and how different models of care and advice compare to structured education programmes already offered to women with type 1 and type 2 diabetes.10 Retinal assessment in the preconception period The evidence in relation to retinal assessment in the preconception period and the GDG’s interpretation of the evidence are presented in Section 5. and format (for example. Specifically it should evaluate different forms of content (i.73 m². in line with the UK National Screening Committee’s recommendations for annual mydriatic two-field digital photographic screening as part of a systematic screening programme.Preconception care diabetes. whether group sessions are more clinically and costeffective than providing care and advice for each woman separately). whether one long session is more clinically and cost-effective than a series of shorter sessions). 59 . Recommendations for renal assessment in the preconception period Women with diabetes should be offered a renal assessment. frequency and timing of contact with healthcare professionals (for example. what topics are covered). which healthcare professionals should be involved (for example. before discontinuing contraception. including a measure of microalbuminuria.11 Renal assessment in the preconception period The evidence in relation to renal assessment in the preconception period and the GDG’s interpretation of the evidence are presented in Section 5. Women with diabetes who are planning to become pregnant should be advised to defer rapid optimisation of glycaemic control until after retinal assessment and treatment have been completed. Retinal assessment should be carried out by digital imaging with mydriasis using tropicamide. If serum creatinine is abnormal (120 micromol/litre or more). There were no research recommendations in relation to retinal assessment in the preconception period. 3. Recommendations for retinal assessment in the preconception period Women with diabetes seeking preconception care should be offered retinal assessment at their first appointment (unless an annual retinal assessment has occurred within the previous 6 months) and annually thereafter if no diabetic retinopathy is found.e. whether preconception care and advice provided by a multidisciplinary team is more clinically and cost-effective than contact with one healthcare professional).4.

1 (95% CI 1. uncontrolled observational study128 that gave all women (n = 1185) a 75 g OGTT found 39. especially around the chest. minority ethnic background. the relative risks (RRs) (with white ethnicity as the reference category) were as follows: black 3. increased weight gain in early adulthood and current smoker.127 In the one UK study included in the HTA systematic review. had higher BMI (27.4% versus 15.3 (95% CI 6. perinatal death.5% (29/1977).5% (20/572) and Indian 4. which increases the risk of shoulder dystocia.4% (54/1218).124 The ‘gold standard’ diagnostic test for gestational diabetes is the 75 g oral glucose tolerance test (OGTT) conducted at 24–28 weeks of gestation. Women with gestational diabetes were significantly older (32. The effects of fetal hyperinsulinaemia include: • an overgrowth of insulin-sensitive tissues such as adipose tissues. One non-randomised.125 1. P < 0. advanced maternal age.8).7 kg/m² versus 23. although in only 16 studies were all women given a diagnostic OGTT regardless of screening result.31 Description of the evidence A systematic review of screening for gestational diabetes was undertaken for a health technology assessment (HTA).3%.4% (26/6135).122 maternal hyperglycaemia results in excess transfer of glucose to the fetus resulting in fetal hyperinsulinaemia.8 kg/m².0001).1) and Indian 11.3 versus 28. After adjusting for age. black 1. P < 0. south-east Asian 7. Rates of gestational diabetes by ethnicity were: white 0. The WHO definition of gestational diabetes encompasses both impaired glucose tolerance (IGT) (fasting blood glucose (FBG) less than 7. shoulders and abdomen.4 4. [EL = 2++] Ethnic groups with increased risk of developing gestational diabetes are South Asian (specifically women whose country of family origin is India.8 to 5. hyperbilirubinaemia and renal vein thrombosis123 • an increased long-term risk of obesity and diabetes in the child. south-east Asian 3.3 years.5% (170/11205) of women were diagnosed with gestational diabetes. fetal polycythaemia. Pakistan or Bangladesh) and black Carribean.8 mmol/litre or more) and diabetes (FBG 7. The potential benefits of recognising and treating gestational diabetes include reductions in ill health in the woman and/or the baby during or immediately after pregnancy.001).124 The review found risk factors for gestational diabetes were obesity. The HTA systematic review found that using risk factors alone as a screening test produced low sensitivities (50–69%) and specificities (58–68%. P < 0.0 mmol/litre and a 2 hour blood glucose 7. BMI and parity.0 mmol/ litre or more or 2 hour blood glucose 11. as well as the benefits of reducing the risk of progression to type 2 diabetes in the longer term and/or future pregnancies being complicated by pre-existing or gestational diabetes.124 which included one UK study.1 to 14.1 mmol/litre or more). In this study women with no risk factors had a gestational diabetes 60 .001) and were more likely to be from a minority ethnic group (55.8 to 18. birth trauma and the need for caesarean section • neonatal metabolic complications such as hypoglycaemia • a hypoxaemic state in utero which may increase the risk of intrauterine fetal death.125 An additional study was identified that was published after the systematic review was published.5).1 Gestational diabetes Risk factors for gestational diabetes Gestational diabetes is defined by the World Health Organization (WHO) as ‘carbohydrate intolerance resulting in hyperglycaemia of variable severity with onset or first recognition during pregnancy’. family history of diabetes.126 The systematic review included 135 studies.6 (95% CI 4.31 According to the Pedersen hypothesis. eight studies).2% (31) of women with gestational diabetes had no risk factors and would have been missed if only selective testing was used.

126 Seventy-four percent (3616/4918) of women agreed to the OGTT.7). hydramnios. then a 3 hour 75 g OGTT was offered. 95% CI 2. The following risk factors were assessed for all women: age and gestational age at entry into the study. Women were also asked for 61 . The results showed a statistically significant difference in risk factors between the two groups. previous macrosomic baby or previous gestational diabetes. The detection rate for gestational diabetes using universal screening was significantly higher than using risk factor screening (2. postnatal and neonatal outcomes. BMI. Other risk factors were family history of diabetes (13/61. [EL = 2+] A cohort study130 examined the effects of pre-pregnancy BMI on antenatal.7% (61) had gestational diabetes.0) and previous macrosomic baby (9/61. 1.7%. 95% CI 1.6. If the capillary FBG measurements were 4. No single risk factor seemed the best indicator for gestational diabetes. If capillary FBG values were 6. [EL = 2++] One study was identified that had been published since the HTA and had given all women a diagnostic OGTT. including the onset of early postpartum diabetes. [EL = 2+] An RCT conducted in the USA132 compared a risk factor-based screening programme for gestational diabetes with universal screening.11). 95% CI 1. Women were tested at 24– 28 weeks using 1 hour 59 g GCT without regard to their last meal. 95% CI 11. OR 23. [EL = 2−] A cross-sectional 5 year investigation conducted in the Netherlands134 examined the clinical utility of antenatal clinical characteristics and measures of glucose tolerance in multi-ethnic women with gestational diabetes for their ability to predict type 2 diabetes within 6 months of birth (early postpartum diabetes). macrosomia in current pregnancy 40% and polyhydramnios in current pregnancy 40%. and pregnancy outcome. previous macosomic baby (more than 4500 g) 12.85).2%.37. OR 2.13). The strongest predictor of developing gestational diabetes was glycosuria (OR 9. The study compared women with gestational diabetes to women with normal glucose tolerance. The most frequent prescreening risk factors were BMI 27 kg/m² or more (present in 65% of women with gestational diabetes) and age 35 years or older (present in 16% of women with gestational diabetes). [EL = 2−] A cohort study135 of 6214 pregnancies among 6034 women evaluated the sensitivity and costeffectiveness of various screening schemes for gestational diabetes.74.08).Gestational diabetes prevalence of 4.45%). OR 3.45 to 7. obstetric and clinical history. BMI 30 kg/m² or more (11/61. CI 1. non-Nordic origin (13/61.33.18 to 4.04. The study showed that apart from family history of diabetes no other risk factor showed an association with the development of early postpartum diabetes.7% vs 1. This prospective population-based study conducted in Sweden offered all pregnant women without diabetes a 75 g OGTT at 28–32 weeks of gestation. intrapartum. The study showed that women who are obese were more likely to develop gestational diabetes (P < 0. Women who did not take the OGTT were more likely to be multiparous and of non-Nordic origin but were less likely to have a family history of diabetes. Women with gestational diabetes had a significantly higher rate of stillbirth.59.8 mmol/litre or more. ethnic group and family history) gave most of the information and that adding other items added little. It was concluded that pre-pregnancy obesity is a risk factor gestational diabetes. OR 5. OR 3. glycosuria in current pregnancy 50%. pre-pregnancy BMI.68 to 11.19. Women in the risk factor group were given a 3 hour 100 g OGTT at 32 weeks of gestation if any risk factor was present. first-degree relative with type 1 diabetes 15%.6 to 63. Of the women who had the OGTT. One study129 found that four risk factors (age.7 mmol/litre.7 mmol/ litre or more. OR 2. [EL = 2+] A study conducted in Denmark133 retrospectively investigated the power of pre-screening based on risk factors to identify gestational diabetes and screening to predict adverse clinical outcomes.56 to 7.001). OR 2.36 to 5. gestational hypertension. The study reported the following PPVs for risk factors: first-degree relative with type 2 diabetes 6. The risk factors with the strongest association were previous gestational diabetes (12/61. 95% CI 1.6 to 48.7). ethnicity. Pregnant women with at least one risk factor were offered capillary FBG at 20 and 32 weeks of gestation.1 mmol/litre or more and less than 6. 95% CI 2.65.47 to 5.8%. the woman was diagnosed as having gestational diabetes. Women in the universal screening group were given a 50 g GCT followed by a 3 hour 100 g OGTT if the plasma glucose at 1 hour was 7. [EL = 2+] A prospective cohort study131 examined the incidence and outcomes of pregnancy in women with gestational diabetes in an Iranian population. 95% CI 1.14) and age 25 years or more (55/61. weight 90 kg or more (8/61. macrosomia and caesarean section.

[EL = 2+] A cross-sectional survey136 of 14 613 women without previous gestational diabetes or other known diabetes who reported singleton pregnancy between 1990 and 1994 were used to measure risk factors for gestational diabetes. whereas a similar audit conducted in 2001 showed that only 69% of healthcare professionals knew when to screen. 10% of women with gestational diabetes. Pre-pregnancy physical activity was not associated with risk for gestational diabetes.001).Diabetes in pregnancy their age and risk factors. Maternal age over 40 years had a two-fold increased risk of gestational diabetes compared with women aged 25–29 years. Oman. 95% CI 1.5 kg.14 to 1. Egypt. Other risk factors. However. [EL = 3] A recent systematic review137 examined the rates and factors associated with recurrence of gestational diabetes among women with a history of gestational diabetes. OGTT results and insulin use. such as maternal age.56.67. 95% CI 1.139 that reported the probability of gestational diabetes given insulin-treated gestational diabetes in a previous pregnancy to be 75–77%. Gestational diabetes developed in 722 women during the study period. Hispanic or Asian ethnicity all had significantly increased age-adjusted RRs for gestational diabetes compared with white women.140 The audit showed that 89% of healthcare professionals were aware of the scheduled time for OGTT administration (26–28 weeks of gestation) in 2002. Higher pre-pregnancy (1989) BMI. Two percent of pregnancies (n = 125) were complicated by gestational diabetes. although the racial groups were not always clearly described in the original studies. or random blood glucose more than 7. see www.eatlas. The most consistent predictor of future recurrence appeared to be non-white race/ethnicity.2–7. of these. the systematic review included two studies138. Risk for gestational diabetes increased directly with greater weight gain (RR 3. inconsistently predicted recurrence of gestational diabetes across studies. Gestational diabetes risk increased with weight gain between age 18 and the year the study began. Family history of diabetes in a first-degree relative increased the risk of gestational diabetes (RR 1. parity.8 mmol/litre had been used.43.80). Syria.39 to 2.0 mmol/litre. No other risk factors were consistently associated with recurrence of gestational diabetes across studies. 70 (56%) were younger than 30 years. The following risk factors and clinical measurements were listed on the appointment form as indications for OGTT: [EL = 3] • • • • • • • • • • • 62 glycosuria ≥ 1+ on more than one occasion or ≥ 2+ on one occasion macrosomia in current pregnancy previous large infant (more than 4. A total of 13 studies were included. An atlas showing the worldwide prevalence of type 2 diabetes in 2007 indicated that Middle Eastern countries (Saudi Arabia.69) for weight gain of 20 kg or more since age 18 years. Jordan. 95% CI 1. Women of African-American. Of women with gestational diabetes. 95% CI 2. Recurrence rates varied between 30% and 84% after the index pregnancy.org/webdata/docs/Map%201. higher BMI at age 18 years and weight gain between 18 years and 1989 all significantly increased the risk for gestational diabetes. [EL = 2++] An audit conducted in the UK in 2002 assessed whether the introduction of an appointment form for administering an OGTT had improved the understanding of antenatal care staff. Gestational diabetes increased with maternal age (P < 0. If a threshold of more than 7.68. Past smokers had no increased risk. United Arab Emirates. Current smoking increased the risk of gestational diabetes when compared with never smokers (RR 1.jpg). 58% had one or more risk factors.05 compared with stable weight). The review showed the recurrence rate of glucose intolerance during subsequent pregnancies varied markedly across studies. Iraq.1_large.8 mmol/litre.70 to 4. . Kuwait and the Lebanon) have high prevalence (more than 10% of women.9 kg 1. Recurrence rates were higher in the minority ethnic populations (52–69%) compared with lower rates found in non-Hispanic white populations (30–37%). Crude RR for gestational diabetes increased 4% (95% CI 2% to 6%) with each year over 25. Qatar.37 to 2. who had screening values of 7. BMI. would have been missed. 1989 (RR for weight gain 5–9.idf.04). [EL = 2+] Additional information about the prevalence of gestational diabetes was obtained using type 2 diabetes as a marker for gestational diabetes. or above the 95th centile for gestational age) previous gestational diabetes first-degree relative with diabetes previous unexpected perinatal death history of polycystic ovary syndrome obesity (BMI more than 30 kg/m²) or booking weight more than 100 kg polyhydramnios Asian ethnic background FBG more than 6.0 mmol/litre.

the GDG also noted that there is a strong possibility of developing frank type 1 diabetes or type 2 diabetes given gestational diabetes in a previous pregnancy (see Section 8. 63 . United Arab Emirates. United Arab Emirates. Kuwait. Syria. Oman. the probability of gestational diabetes given insulin-treated gestational diabetes in a previous pregnancy is approximately 75%. Pakistan or Bangladesh) – black Caribbean – Middle Eastern (specifically women whose country of family origin is Saudi Arabia. Oman.2. From evidence to recommendations The following have been shown to be independent risk factors for gestational diabetes and should be recognised as such by healthcare professionals: • • • • • BMI more than 30 kg/m² previous macrosomic baby weighing 4.Gestational diabetes Evidence statement Evidence shows that risk factors for developing gestational diabetes are: pre-pregnancy obesity. Iraq.5 kg or above previous gestational diabetes family history of diabetes (first-degree relative with diabetes) family origin with a high prevalence of diabetes: – South Asian (specifically women whose country of family origin is India. The probability of gestational diabetes given insulin-treated gestational diabetes in a previous pregnancy is approximately 75%. Kuwait. Pakistan or Bangladesh). Lebanon or Egypt). advanced maternal age. Qatar. Lebanon or Egypt). minority ethnic background. Jordan. Family origins with a high prevalence of gestational diabetes are South Asian (specifically women whose country of family origin is India. United Arab Emirates. previous macrosomic baby (4500 g or more for white and black women). Syria. Recommendations for risk factors for gestational diabetes Healthcare professionals should be aware that the following have been shown to be independent risk factors for gestational diabetes: • • • • • body mass index above 30 kg/m² previous macrosomic baby weighing 4. black Carribean and Middle Eastern (specifically women whose country of family origin is Saudi Arabia.2). Qatar. Recurrence rates for gestational diabetes varied from 30% to 84% after the index pregnancy. and current smoker. Syria. Recommendations relating to recurrence of gestational diabetes and the need for early testing for gestational diabetes in future pregnancies are presented in Section 8. Moreover. Jordan. In its discussions.3). There were no research recommendations relating to which women are at high risk of gestational diabetes. Pakistan or Bangladesh) – black Caribbean – Middle Eastern (specifically women whose country of family origin is Saudi Arabia. The consensus view of the combined GDGs for antenatal care and diabetes in pregnancy is that advanced maternal age should not be used as a risk factor for gestational diabetes because this would result in most pregnant women receiving an OGTT (see Section 4. Qatar. Jordan. Kuwait. Iraq. The probability of gestational diabetes for a woman who has had gestational diabetes in a previous pregnancy is 30–84%. previous gestational diabetes.5 kg or more previous gestational diabetes family history of diabetes (first-degree relative with diabetes) family origin with a high prevalence of diabetes: – South Asian (specifically women whose country of family origin is India. Lebanon or Egypt). and so it is straightforward to demonstrate the cost-effectiveness of offering a diagnostic test for gestational diabetes to women who have had gestational diabetes in a previous pregnancy. family history of diabetes. Oman. Iraq. increased maternal weight gain in early adulthood.

1 mmol/litre ≥ 6. ROC curves for the prediction of neonatal cranial : thoracic ratio ≤ 10th percentile identified a threshold 1 hour value of 8.141 Diagnostic tests vary in the glucose load to be used.7 mmol/litre Plasma venous ≥ 7.6–9.1 World Health Organization criteria for the 75 g OGTT31 Whole blood venous Whole blood capillary Fasting ≥ 6.6 mmol/litre and group C (n = 40) had 2 hour values of 6.145 [EL = 2++] A cohort study considered the association between glucose tolerance and the ratio of fetal cranial : thoracic circumference in 829 women who had been diagnosed with gestational diabetes (Carpenter and Coustan criteria). It found progressively increasing plasma glucose associated with increased incidence of macrosomia and caesarean section.6 mmol/litre ≥ 7.0 mmol/litre ≥ 8.83).3 mmol/litre at 16–20 weeks of gestation and 8.8 mmol/litre at 26–30 weeks of gestation.81.9 mmol/litre Table 4.8 mmol/litre 2 hours ≥ 8.6–6.2 Criteria for the 100 g OGTT (gestational diabetes is diagnosed if two or more measurements in one column are abnormal.2).15 to 2.1 mmol/litre ≥ 8. group B (n = 58) had 2 hour levels of 5.0 mmol/litre ≥ 8.8 mmol/litre ≥ 10.3 mmol/litre ≥ 10. [EL = 2++] The third study was of 3352 women who did not fulfil the National Diabetes Data Group (NDDG) criteria for gestational diabetes.1 mmol/litre. 1 hour and 2 hour values. The study found a positive correlation between the 2 hour glucose level and macrosomia and maternal complications (pre-eclampsia and/or caesarean section.146 The study found the 1 hour value on a 75 g glucose load was significantly associated with a cranial : thoracic ratio of ≤ 10th percentile.144 This study found a progressive increase in the prevalence of macrosomia with incremental fasting. The most commonly used tests are the 2 hour 75 g OGTT used by the WHO (Table 4.1 mmol/litre ≥ 5.9 mmol/litre Description of the evidence Levels of glucose intolerance and adverse pregnancy outcomes Three cohort studies found adverse pregnancy outcomes increased with increasing gradients of glucose intolerance.0 mmol/litre ≥ 9.0 mmol/litre ≥ 7.0 mmol/litre ≥ 9.0 mmol/litre 3 hours ≥ 6. Receiver operating characteristic (ROC) curves were constructed but no inflexion point could be identified. In all three studies the glucose intolerance was previously undiagnosed and untreated and the confounding effects of obesity and maternal age were controlled. i.Diabetes in pregnancy 4.6 mmol/litre.8 mmol/litre 2 hours ≥ 6. P < 0.8 mmol/litre and 2 hour values less than 11. 95% CI 1.01).0 mmol/litre.0 mmol/litre ≥ 5.2 Diagnosis of gestational diabetes Internationally there are at least six different criteria for the diagnosis of gestational diabetes. test for linear trend.1 mmol/litre ≥ 7.e. if they exceed the levels indicated)142 Time Fasting 1 hour O’Sullivan and Mahan National Diabetes Data Group (NDDG) Carpenter and Coustan (whole blood) recommendations (plasma glucose) adaptation (plasma glucose) ≥ 5. This was apparent as early as 16–20 weeks of gestation (adjusted OR 1.8 mmol/litre Plasma capillary ≥ 7.1)31 and the 3 hour 100 g OGTT as recommended by the American Diabetes Association (Table 4. [EL = 2++] The second study was of 3352 women with 75 g OGTT FBG less than 5. [EL = 2++] 64 . the timing and the type of blood sample. The first study143 divided women with normal results on a 100 g OGTT (O’Sullivan and Mahan criteria) into three groups: group A (n = 151) had 2 hour glucose levels below 5.142 Table 4.

8 and 8. All four studies controlled for the effect of potential confounders including obesity and age. P < 0. The glucose profiles of the gestational diabetes group before treatment were identical to the profiles of the group with one abnormal value. Neonatal metabolic disorders were significantly higher in the group with one abnormal value (15%) when compared with the gestational diabetes and normal groups (3%. Logistic regression found a significant increase in birthweight in women with IGT compared with women with normal glucose tolerance.0– 11. The control group was 220 women without diabetes and with amniotic fluid insulin levels less than 42 pmol/litre. P < 0. [EL = 2++] The fourth study was a cohort study that included 1190 black women and 865 white women at 24–29 weeks of gestation. After the gestational diabetes group underwent treatment there was a significant difference in mean blood glucose levels and blood glucose profiles between the gestational diabetes group and the group with one abnormal value.150 The study found that IGT was significantly and independently associated with an increased incidence of caesarean section.8 mmol/litre or more had a sensitivity of 54%. The first study was a case–control study which compared 126 women with gestational diabetes diagnosed according to the NDDG criteria (two abnormal values on the 100 g OGTT) to 126 women with one abnormal value and 126 women with normal values. A fasting level of 5.0 mmol/litre) with 815 controls. among black women. LGA and macrosomic babies and admission to neonatal intensive care for 2 days or longer. In all four studies the IGT was previously undiagnosed and untreated. [EL = 2++] The second study was a cohort study149 of 3260 women that divided women into four groups based on the results of a 2 hour 75 g OGTT: group 1 had a 2 hour value less than 7.01) and the gestational diabetes group (34% versus 12%.0 mmol/litre or more had a sensitivity of 58% and a 2 hour level of 7. and group 4 had a 2 hour level more than 11.8 mmol/litre (n = 2596) and were not treated.7 mmol/litre and 2 hour level of 9. Among white women 70 had gestational diabetes.8–8. The study found that a glucose load test result of 10. The 1 hour OGTT value had the highest sensitivity for raised insulin concentration (97% of women with elevated amniotic fluid insulin concentration had 1 hour levels more than 8. There was no significant difference between the gestational diabetes group after treatment and the normal group.148 There was no significant difference in mean blood glucose values between the gestational diabetes group before treatment and the group with one abnormal OGTT value.5 mmol/litre or more was associated with unfavourable perinatal outcomes and might warrant further follow-up regardless of the subsequent OGTT results. [EL = 2+] A retrospective cohort study152 hypothesised that elevated 50 g glucose load test results could be associated with unfavourable perinatal outcomes even in the absence of an actual diagnosis of gestational diabetes.8 mmol/litre. preterm birth.Gestational diabetes A case–control study considered the association between glucose tolerance and elevated insulin levels in amniotic fluid.9%). [EL 2−] 65 . Among black women 32 had gestational diabetes. 40 had IGT and 1080 had normal glucose tolerance. group 3 had a 2 hour value of 9. group 2 had a 2 hour value of 7. 13 had IGT and 820 had normal glucose tolerance.0–11. [EL = 2++] The third study was a case–control study that compared 213 women with undiagnosed as well as untreated IGT (2 hour 75 g OGGT fasting level less than 6.9 mmol/litre compared with women with 2 hour OGTT levels less than 7. There was no significant difference in LGA and macrosomic babies between the gestational diabetes group and the normal group.147 The study included 220 women with elevated amniotic fluid insulin levels (42 pmol/litre or more) who had an OGTT at 28 weeks of gestation (161 had a glucose load of 1 g/kg and 59 had 75 g). [EL = 2++] Four studies found that women with IGT had more adverse outcomes than women with normal glucose tolerance.9 mmol/litre (n = 289) and were not treated.01).01). P < 0.0 mmol/litre (n = 278) and were treated. There was a significant difference in the incidence of LGA and macrosomic babies between the group with one abnormal value and the normal group (34% versus 9%.151 All women were given a 100 g OGTT (Carpenter and Coustan criteria).1 mmol/litre (n = 97) and were treated. The profiles of the gestational diabetes group before treatment and the group with one abnormal value were significantly different to the group with normal OGTT values. but not white women. The frequency of macrosomia increased by more than 50% in women with 2 hour OGTT values between 7.

8 mmol/litre and 2 hour level of 7.3 mmol/litre at 16–20 weeks of gestation and 8. This finding suggests a possibility of using a 1 hour 75 g glucose load as a single test for the diagnosis of gestational diabetes. such as diet and exercise and self-monitoring of blood glucose.3 for further details).2). in a diagnosis of gestational diabetes. and hypoglycaemic therapy. One study found an independent and significant association between the 1 hour value following a 75 g glucose load and a neonatal cranial : thoracic ratio ≤ 10th percentile. therefore. 4. ROC curves identified a threshold value of 8.153 The rate of serious perinatal outcomes among babies was significantly lower in the intervention group (1% versus 4%. The aim of the study is to establish the level of hyperglycaemia that results in a significant risk of adverse pregnancy outcome. 66 .9 The clinical evidence in relation to treating gestational diabetes is presented in this section.9% at 28 weeks of gestation.3.01). a high-quality RCT has shown that treating gestational diabetes (as defined by the WHO) improves maternal and fetal outcomes. The number needed to treat to prevent a serious outcome in a baby was 34. This is supported by a study that found that 97% of women with elevated amniotic fluid insulin concentration had a 1 hour OGTT value of more than 8. This risk remained even when the woman and physician were blinded to the diagnosis of IGT and after controlling for maternal age and obesity.8–11. [EL = 1++] An international multicentre prospective cohort study (the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study) is due to report soon. P = 0. This supports the inclusion of IGT. diagnosis and treatment was addressed through a single health economics model which is described below. regular insulin and/or rapid-acting insulin analogues (lispro and aspart). The cost-effectiveness of screening.0 mmol/litre) to treatment and 510 women with IGT to routine care. as defined by the WHO. The recommendations in relation to the diagnosis of gestational diabetes are presented in Section 4. There was no significant difference between groups in maternal quality of life (see Section 4. adopted the 2 hour 75 g OGTT using the WHO criteria as the diagnostic test for gestational diabetes administered at 24–28 weeks of gestation. The GDG has.153 Treatments for gestational diabetes include lifestyle interventions. Evidence statement Three studies were identified that demonstrated that adverse outcomes increased with increasing gradients of glucose intolerance. From evidence to recommendations The evidence shows that when gestational diabetes is diagnosed using the 2 hour 75 g OGTT and the WHO criteria there is an increased incidence of (treatable) maternal and neonatal complications. Four studies were identified that found that women with levels of glucose tolerance above normal but below the threshold values for gestational diabetes as defined by the NDDG or Carpenter and Coustan criteria have significantly higher risk of adverse maternal and neonatal outcomes. The clinical evidence in relation to screening methods is presented in the antenatal care guideline. Furthermore. using hypoglycaemic agents (metformin and glibenclamide). diagnosis and treatment for gestational diabetes.8 mmol/litre at 26–30 weeks of gestation.Diabetes in pregnancy Treating impaired glucose tolerance An RCT (the Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS)) allocated 490 women with IGT (2 hour 75 g OGTT fasting level less than 7. Description of the evidence The ACHOIS trial has shown that treating gestational diabetes as defined by the WHO improves outcomes for women and babies (see Section 4.3 Screening and treatment for gestational diabetes This section includes work undertaken jointly with the GDG for the NICE antenatal care guideline update on the cost-effectiveness of screening.

[EL = 1+] A case–control study compared the diets of 16 women with newly diagnosed gestational diabetes with those of 24 healthy pregnant women. In non-pregnant women the mean (SD) 1 hour postprandial value was 7 (1. [EL = 2++] In a crossover study15715 women without diabetes were evaluated following three diets. [EL = 1++] A crossover study of 14 non-pregnant women without diabetes compared low glycaemic index (GI) diets with high GI diets.159 The low GI group had significantly lower FBG (3. Diet 2 consisted of 1820 kcal.02). P < 0.05). Unrefined carbohydrate sources (mostly complex) were used in both diets. Birthweight. fat mass and maternal weight gain were all significantly lower in the low GI group (P < 0. 40 g carbohydrate and 52 g dietary fibre. 17 g protein. 20 g protein. Each meal contained 17% protein. 39 g fat. 45% fat.01). Pregnant women demonstrated improved glucose homeostasis with advancing pregnancy.154 In women with gestational diabetes adverse outcomes have been associated with postprandial hyperglycaemia. 22 g protein.3) mmol/ litre.161 The highly unrefined carbohydrate diet was made up of 70% carbohydrate. The average increase in blood sugar was significantly lower on the low GI diet (P < 0. Fasting free fatty acids levels were 14% lower than on the low carbohydrate diet (P < 0. the mean (SD) 1 hour postprandial value was 6 (1.158 After 7–10 days on the diet postprandial glucose was measured after a 540 kcal test meal. fasting insulin levels (63 nmol/litre versus 131 nmol/litre. Insulin response was also significantly lower (P < 0.01) on the low GI diet. There was a significant improvement in glucose tolerance on the highly unrefined carbohydrate diet (P < 0.05).001). [EL = 1+] An RCT compared ten pregnant women on a low GI diet with ten pregnant women on a high GI diet (none of the women had diabetes). The low carbohydrate diet was made up of 35% carbohydrate. 60 g carbohydrate and 84 g dietary fibre.05). 10% fat.28 mmol/litre. 16 g fat. For example.5) mmol/litre in women in the 13th week of gestation and 5.56 mmol/ litre per minute. [EL = 2+] A small crossover study in five women with gestational diabetes compared a diet that was low in fat and high in carbohydrates of low glycaemic index with a low carbohydrate diet. P < 0. [EL = 1+] A randomised crossover study in seven women with IGT and previous gestational diabetes compared the effect on insulin secretion and insulin sensitivity of low GI/high dietary fibre bread with high GI/low dietary fibre bread. 20% protein and 70 g fibre each day. 28% fat and 55% carbohydrate but differed in the GI of the carbohydrates (54 versus 92).05). The study revealed a loss of insulin sensitivity on diet 1 but not on diets 2 or 3.Gestational diabetes The primary goal of interventions for gestational diabetes is to maintain near normal glycaemic control in order to reduce morbidity and mortality in women and babies. Diet 1 reflected a typical ‘Western’ diet and consisted of 1830 kcal.162 The study comprised two 3 week study periods separated by a 3 week wash-out period.78 versus 4. Diet The goals of dietary interventions in pregnancies complicated by diabetes are the optimisation of glycaemic control while avoiding ketoacidosis and minimising the risk of hypoglycaemia in women taking insulin.01).05) and average insulin increase (186 nmol/litre per minute versus 324 nmol/litre per minute.155 therefore an important aim of dietary therapy is reducing postprandial glucose levels. Diet 3 consisted of 1850 kcal. ponderal index. 41 g carbohydrate and 10 g dietary fibre.156 All women consumed a diet high in carbohydrates of low glycaemic index.160 The women with gestational diabetes consumed significantly fewer carbohydrate foods with low GI values (P < 0. 20% protein and 70 g fibre each day.05).4) mmol/litre in women at 37–40 weeks of gestation.9 (1. Glucose tolerance was assessed by measuring plasma glucose and insulin response to a 50 g glucose load. Following inclusion of low GI/high dietary fibre bread 67 . P < 0. Fasting plasma cholesterol concentrations on the highly unrefined carbohydrate diet were 6% lower than on the low carbohydrate diet (P < 0. 38 g fat. Seven studies were identified that considered the influence of a diet based on carbohydrates of low glycaemic index on postprandial glucose levels. A study in an African population compared blood glucose values in women during pregnancy with those in non-pregnant women. P < 0.60 mmol/litre per minute versus 1. average postprandial glucose increase (0.

The criterion for remaining on dietary treatment alone was a diurnal glucose profile averaged over six measurements of 6 mmol/litre or less. BMI and parity with a negative GCT result (screen negative). high-density lipoprotein cholesterol or triglycerides. All women had excellent glycaemic control (2 hour post-breakfast plasma glucose levels less than 7 mmol/litre).166 The women who were obese were prescribed a diet of 1700–1800 kcal/ day (two women received small amounts of regular insulin before each meal). [EL = 2−] Calorie restriction Three observational studies were identified that considered calorie restriction in women with gestational diabetes.165 The control group was contaminated. HbA1c at 35 weeks of gestation was 6. The women who were lean were prescribed 2000–3000 kcal/day and the women with normal glucose tolerance were prescribed a diet appropriate for pregnancy. [EL = 2+] A cohort study compared 22 women with gestational diabetes who were obese (BMI 27 kg/m² or more) to 31 women with gestational diabetes who were lean (BMI less than 27 kg/m²). Two groups were compared: group 1 consisted of women who were obese with previous gestational diabetes and group 2 consisted of women who were obese with no previous history of gestational diabetes. one during the day and one overnight. The total calorie intake was prescribed to be 30% less than consumed before pregnancy. [EL = 2+] A prospective randomised crossover study167 studied 40% versus 55% carbohydrate calorically restricted diets to compare weight loss and metabolic response. age.Diabetes in pregnancy all women lowered their insulin response to the intravenous glucose challenge (mean 35%). (ii) a group of 35 women matched on ethnic background. The study showed that a weight-loss regimen consisting of 40% carbohydrate resulted in lower triglyceride levels than was achieved with a 55% carbohydrate content diet in women who were obese. but women in the intervention group reduced their pregnancy weight gain without significant ketonuria or fetal or maternal compromise. Women with gestational diabetes were prescribed a calorie-restricted diet with complex carbohydrates replacing refined carbohydrates and fats. protein as fish.4 ± 1. and analysed hormones and metabolites involved in intermediary metabolism.7% in the women who were lean and 6. [EL = 1+] A cohort study163 compared glucose measurements for women with diet-controlled gestational diabetes during two 9 hour fasts.4% in the women who were obese. An observational study included 35 women with gestational diabetes on a calorie-restricted diet. 68 . white meat.8 ± 0.5% of the intervention group required insulin and 28. Mean birthweight was significantly higher in women with gestational diabetes who were obese than in the other two groups despite a lower caloric intake and a smaller weight gain during pregnancy.8% of babies were LGA. Ketonuria was not observed in any woman even when reported caloric intakes were as low as 1500–1600 kcal/day. The composition of the diet was 50% carbohydrate. the screen positive group had a significantly higher incidence of macrosomia than the women with gestational diabetes and the general antenatal population (P < 0. insulin. No changes were found in fasting levels of glucose.164 The study used three control groups: (i) 2337 healthy pregnant women (general antenatal population). The women in the gestational diabetes group had significantly less weight gain during pregnancy than the control groups (P < 0.005). [EL = 2++] An RCT was designed to compare the effect of a moderate reduction in energy intake (30%) in women with gestational diabetes who were obese on requirement for insulin therapy and incidence of macrosomia. 15% protein and 35% fat. [EL = 1+] Exercise Three RCTs and a retrospective study were identified that compared diet with diet plus exercise in women with gestational diabetes.2% in the control group. and fat as polyunsaturated fat. beans and dairy products. A control group consisted of ten women who were not obese (BMI 24 kg/m² or less) and with normal glucose tolerance.005).7 ± 0. The study showed that glucose concentrations in women with diet-treated gestational diabetes were significantly higher 3–9 hours after an evening meal. In contrast. with most carbohydrate taken as unrefined carbohydrate. 6. and (iii) a matched group of 35 women with a positive screening test result but a negative diagnostic test result (screen positive). Glycaemic control was good. 17. Incidence of macrosomia was similar to that of the general antenatal population.

168 Exercise consisted of arm movements while seated in a chair. When to initiate insulin Insulin therapy is needed when near-normal blood glucose control cannot be achieved by diet alone. The glucose challenge fasting value was 3. Within the intervention group. [EL = 1++] The second RCT compared 16 women with gestational diabetes treated with diet alone to 16 women treated with diet plus resistance exercise three times a week. although women exercised more during the postnatal period than before or during pregnancy. Among 68 women successfully treated for a minimum of 6 weeks. 15 women (14%) were switched to diet plus insulin. The women’s husbands/partners strongly influenced their levels of exercise during pregnancy and in the postnatal period.Gestational diabetes In the first RCT 20 women with gestational diabetes were randomised to diet or diet plus 20 minutes of supervised exercise three times a week. Otherwise there were no differences between the two groups in maternal blood glucose control. There were no significant differences in gestational age at birth. By the 6th week women in the diet plus exercise group had mean HbA1c of 4. HbA1 at birth.3 mmol/litre (P < 0. 30% of the women who exercised two to three times per week were prescribed insulin therapy compared with 67% of those who exercised less than twice per week.5–5. The duration of dietary treatment prior to initiation of insulin will depend on gestational age at diagnosis and the level of glycaemic control.05). aiming at fasting and postprandial values below 5 and 6. [EL = 1++] The third RCT was a randomised crossover study170 which compared fasting glucose and insulin levels. a subgroup analysis that looked only at women with a pre-pregnancy BMI more than 25 kg/m² found that women in the diet plus exercise group had a significantly lower incidence of insulin use (3/10 versus 8/10.173 Blood glucose levels were evaluated weekly in a high-risk clinic where medical and nutritional support and counselling were provided. skinfold thickness or C-peptide concentration in cord serum. They found that a single bout of exercise did not blunt the glycaemic response observed following a mixed nutrient meal. obstetric or neonatal complications. and incremental area of the glycaemic and insulin curves following a mixed nutrient meal with or without an exercise stress in women with normal glucose levels and women with gestational diabetes.4 ± 0.172 Selfmonitoring of blood glucose was performed six times per day three times per week. [EL 1+] A retrospective study171 sought to examine the postnatal exercise beliefs and behaviours of women who had been diagnosed with gestational diabetes in a recent pregnancy. The most common barrier to exercise was lack of time.2% in the diet-only group (P < 0. when all postprandial results were pooled.001).8 mmol/ litre (see Section 5.7 ± 0.2). P < 0. peak glucose and insulin levels.1).9 ± 0. The study showed that the perceived advantage of exercise during pregnancy was controlling blood glucose and during that postnatal period it was controlling weight.0 mmol/litre versus 10.9 ± 1.2% compared with 4.4 mmol/litre versus 4. An RCT randomised 202 women with IGT to treatment with diet or diet plus insulin. [EL = 2−] Blood glucose monitoring Blood glucose targets during pregnancy for women with diabetes (including gestational diabetes) are a preprandial value of 3. However. however.8 ± 0. There were no differences in self-monitored blood glucose values.001).001. The 1 hour postprandial value was 5. Insulin reduced birthweights significantly in 69 . rate of caesarean section or birthweight.169 The number of women who required insulin treatment was not significantly different between the two groups.2 ± 0. the mean birthweight and macrosomia rate were reduced significantly in the insulin-treated group. P < 0.9 mmol and a 1 hour postprandial value of less than 7.05). respectively. Insulin doses were adjusted according to blood glucose values. In the diet-only group.16 mmol/litre. The recommendations for self-monitoring of blood glucose during pregnancy are to test FBG levels and to test blood glucose 1 hour after eating (see Section 5. respectively.5 mmol/litre. women in the diet plus exercise group had significantly lower values (P < 0. [EL = 1++] Another RCT randomised 108 women with gestational diabetes to receive either diet alone or diet plus insulin for glycaemic control. birthweight. The amount of insulin that was prescribed was significantly lower in the diet plus exercise group as was the latency period between the first clinic visit and the initiation of insulin. Women in the diet-only group were switched to diet plus insulin if fasting and postprandial values exceeded 7 and 9 mmol/litre.

If the fasting and postprandial levels were persistently less than 5. Pre-pregnancy BMI more than 28 kg/m². P < 0.95 mmol/litre or less and 2 hour postprandial values 6. shoulder dystocia (16.6 mmol/litre (n = 2) or because of inadequate self-monitoring of blood glucose 70 .177 All women had an ultrasound scan at 29–33 weeks of gestation. P < 0. P < 0. An additional nine women in the intervention group began insulin subsequently because of an abdominal circumference ≥ 70th percentile (n = 6) or FPG more than 6. There were no differences between the study and reference populations in mean birthweight or incidence of macrosomia.01 for diet only. [EL = 1+] An RCT compared a management strategy based on glycaemic targets (control group.7 mmol/litre. Women in the two treatment groups underwent breakfast tests at baseline and at 36–37 weeks of gestation.3–0.3 episodes per week. early diagnosis of gestational diabetes. respectively. Women in the intervention group were prescribed insulin if fetal abdominal circumference was ≥ 70th percentile or fasting plasma glucose (FPG) more than 6.001) compared with women with a fetal abdominal circumference ≤ 90th percentile. [EL = 2+] Ultrasound measurement of abdominal circumference A cohort study included 201 women with gestational diabetes who underwent ultrasound at 30–33 weeks of gestation. The insulin-treated group had much higher caesarean section rates (43%. The sensitivity of an abdominal circumference measurement at 30–33 weeks of gestation for detecting macrosomia was 88%.4 mmol/litre or less and 2 hour postprandial values of 6. n = 48) with one based on ultrasound evaluation (intervention group.6 mmol/litre. although they were 0. respectively.02) were lower in the insulin group.05 versus other groups). blood glucose could not be controlled below these levels insulin therapy was started. In the intervention group. [EL = 2++] A cohort study compared 52 women with gestational diabetes who were treated with diet alone with 23 who required insulin.2 mmol/ litre and 6.8 mmol/litre and 7. Eleven women (7. 7. the PPV was 56% and the NPV was 96%.3 mmol/litre) in the insulin group was 0. early rise in triglyceride levels and elevated 3 hour OGTT levels were predictive of insulin treatment. n = 48) for women with gestational diabetes.005).6%. adjustments were made to the diet. P < 0. Weekly capillary blood glucose (CBG) readings also reduced significantly in both treatment groups. P < 0. If self-monitored blood glucose was consistently equal to or higher than these values. There were no episodes requiring assistance. the specificity was 83%.7% versus 0.174 A group of 2153 healthy pregnant women served as controls.9 mmol/litre.176 Women with a fetal abdominal circumference > 90th percentile had a significantly higher incidence of caesarean section for failure to progress in labour (41.02) and the incidence of LGA babies (13% versus 45%. after dietary adjustments. 21 were immediately started on insulin. Insulin dose was adjusted to achieve preprandial CBG 4.7% versus 14. [EL = 2+] An RCT involved 303 women with gestational diabetes. women continued with the diet and reduced monitoring to 1 day per week.7%. Birthweights (3647 ± 67 g versus 3878 ± 84 g.6 mmol/litre or less.3%. The women with an abdominal circumference ≥ 75th percentile were randomised to diet (n = 29) or diet plus insulin (n = 30).6 mmol/litre lower in the insulin group throughout the treatment period (P < 0. The frequency of symptomatic hypoglycaemia (less than 3. women with gestational diabetes used monitors to measure fasting and 2 hour postprandial levels three to five times per week. During the first 1–2 weeks following diagnosis.001). P < 0. Of these.175 The aim of the study was to identify which factors predicted insulin treatment.8% versus 3.178 Women in the control group were prescribed insulin immediately. Women with poor glycaemic control were at greatest risk (30%) for fetal overgrowth whether initially receiving insulin or not. Glucose levels in both groups were lower at the follow-up breakfast test than at baseline (P < 0.2%) required insulin. Sixty-eight percent (205/303) had abdominal circumference < 75th percentile.Diabetes in pregnancy women receiving insulin.05 mmol/litre or less. the dose being adjusted to achieve preprandial CBG concentrations 4.005). [EL = 1+] A cohort study included 153 women with gestational diabetes treated with diet and intensive self-monitoring of blood glucose. Of these.8 mol/litre or more. 45% (22) of women had a fetal abdominal circumference of ≥ 70th percentile at enrolment.0001 for insulin group).001) and birth trauma (20. P < 0. If.3% (15) were placed on insulin therapy because of persistent fasting glycaemia 5. as were neonatal skinfold measurements (P < 0. Women in this study consumed a diet containing 25 kcal/kg/day and started insulin when fasting and 2 hour postprandial blood glucose levels were higher than 5.

when fasting capillary glucose (FCG) less than 4. Blood glucose targets for the intervention group were 4.Gestational diabetes (n = 1). Insulin was not prescribed.7 mmol/litre postprandial if < 75th abdominal circumference centile.03). and 5.05 mmol/litre.5%.9 mmol/litre in the glibenclamide group and 5. LGA babies (12% versus 10%). The intervention group had significantly lower incidence of macrosomia (3.3% versus 8. There were no significant differences between the glibenclamide and insulin groups in the percentage of babies who were LGA (12% 71 .6 mmol/litre postprandial. 2 hour more than 6.9 ± 0. Glucose targets were not discussed with the women and glucose values were not used as a guide to management unless any fasting value was more than 6.180 Ultrasound examinations were scheduled every 2 weeks in the intervention group and at 34–38 weeks of gestation in the control group. All women were Medicaid recipients. Birthweights (3271 ± 458 g versus 3369 ± 461 g).6 mmol/ litre or any 2 hour value was more than 11. In the control group insulin was prescribed if two glucose profiles had two or more elevated values (FCG more than 4.95 mmol/litre and 2 hour postprandial less than 6.9%.6 mmol/litre) or if four profiles had at least one elevated value during a 2 week period. LGA babies (6.43% versus 33.05). In the intervention group.3%) and neonatal morbidity (25% versus 25%) did not differ between groups. Insulin dose was adjusted to achieve FCG less than 4. [EL = 1+] An RCT compared 151 women with gestational diabetes managed according to ultrasound assessment of abdominal circumference with 78 women in a control group who were conventionally managed (self-monitoring of blood glucose plus diet or self-monitoring plus diet and insulin). One RCT74 enrolled 404 women with gestational diabetes. P < 0. [EL = 1++] Oral hypoglycaemic agents Two RCTs and four cohort studies were identified that considered the use of glibenclamide in women with gestational diabetes. Insulin treatment was required in 16.181 Seventy-three of the women were evaluated at both 28 weeks of and 32 weeks of gestation and 68 women at 32 weeks of gestation only. Approximately 83% were Hispanic.5 mmol/litre postprandial if ≥ 75th abdominal circumference centile. Insulin dose was adjusted to achieve FCG less than 4. P < 0.0 mmol/litre in the insulin group (P = 0.4 mmol/litre fasting and 5.6 mmol/litre.7% of the control group and 30. Women were randomly assigned at 11–33 weeks of gestation to receive either glibenclamide (n = 201) or insulin (n = 203). insulin therapy was started when abdominal circumference of the baby exceeded the 75th percentile. The findings suggest that fetal ultrasound assessment at 28 weeks of gestation and early introduction of insulin therapy should be encouraged since insulin administration after 32 weeks of gestation is less effective in lowering the rate of macrosomic babies.5 mmol/litre fasting and 7. In the control group 30% (30) met the criteria for insulin.99).05) and babies who were LGA (7.4 mmol/litre and 2 hour value was less than 5.0 mmol/litre. babies of women who did not receive insulin had lower birthweights than babies of women treated using insulin (3180 ± 425 versus 3482 ± 451.33%. [EL = 1+] An RCT involving 141 women with gestational diabetes investigated the gestational age considered adequate for ultrasound assessment of abdominal circumference to prevent fetal overgrowth.05).179 In the intervention group insulin was started when fetal abdominal circumference was > 75th percentile (before 36 completed weeks of gestation). Babies of women who had ultrasound assessment only at 32 weeks of gestation had a significantly higher percentage of macrosomic babies compared with women evaluated at 28 weeks of and 32 weeks of gestation (71.4 mmol/litre and 2 hour less than 6.024). Blood glucose targets for the control group were 5 mmol/litre fasting and 6. Eight women (4%) in the glibenclamide group required insulin treatment. The mean blood glucose concentrations during treatment were 5. There was no difference between the intervention and control groups in rates of caesarean section (18% versus 19%).5 mmol/litre.95 mmol/litre. SGA babies (12% versus 13%) or any other neonatal outcome. Ultrasound was performed at entry and then at 4 week intervals. P < 0. P = 0.5% of the intervention group (P = 0. 12% were white and 5% were black.3% versus 11. In the intervention group 40% (40) of women met the criteria for insulin. [EL = 1+] An RCT compared 99 women with gestational diabetes managed according to ultrasound assessment of abdominal circumference (intervention group) to 100 women with gestational diabetes managed according to glycaemic levels (control group).9 ± 1.9% versus 17. regardless of abdominal circumference. In both groups. The caesarean section rate was significantly lower in the control group.

95% CI 1. Neonatal hypoglycaemia was observed in one baby in the insulin group. 16% (12) switched to insulin prior to birth.63).27. [EL = 1++] The other RCT compared 27 women with gestational diabetes assigned to insulin therapy. 95% CI 76. [EL = 2+] A cohort study reported outcomes for 197 women with gestational diabetes.073).32. gestational age at diagnosis. Forty-nine percent (36/73) required caesarean section. [EL = 2++] A cohort study reported on 75 women with gestational diabetes treated using glibenclamide. In the glibenclamide group. There were no significant differences between the two groups in birthweight. Eight babies in the glibenclamide group had birth injuries (four clavicle fractures.52).182 Whenever maximum dose of oral hypoglycaemic agents was reached without reaching glucose control. one brachial plexus injury and three bone injuries). Glucose control was not achieved in five women in the glibenclamide group (20. In the 59 women whose gestational diabetes was controlled with glibenclamide. Nine women experienced side effects (four malaise and weakness. There were no significant differences in fasting or postprandial glucose levels or in mean birthweights in the three groups. Sixty-three percent (124) of women achieved satisfactory blood glucose control with diet alone and 37% (73) required glibenclamide.1%). ethnicity. Of the 73 women treated using glibenclamide. neonatal hypoglycaemia (9% versus 6%). FPG on OGTT) glibenclamide treatment was significantly more likely to be associated with achieving mean glucose goals (adjusted OR 0. For these women the mean BMI was 31.184 In the study institution.34 to 0. The targets for blood glucose control were mean plasma fasting glucose 5 mmol/litre or less and mean 1 hour value 7. Eighty-four percent (63/75) of women treated using glibenclamide achieved good glycaemic control. one light headedness. No maternal hypoglycaemia requiring hospital admission was reported. Women in the glibenclamide group were significantly more likely to be morbidly obese than the group controlled by diet (P < 0. macrosomic (7% versus 4%).2 mmol/litre or more detected 92% of women who converted to insulin but had a false-positive rate of 70%.2 mmol/litre (P = 0. [EL = 1+] A cohort study compared 236 women treated using glibenclamide with 268 historical controls treated using insulin. 9% had caesarean sections 72 . Only three women who switched for poor control were on the maximum dose of 20 mg/day.5% in the acarbose group (P = 0.6%) had acceptable glucose control on medical therapy. Fasting levels of 5. one baby in the acarbose group and six babies in the glibenclamide group (P = 0. targets for blood glucose for women with gestational diabetes were mean level 5. fasting 5.03). two nausea.17 to 4. 95% CI 0.183 The study population was a large US managed care organisation in which the extremes of socio-economic distribution were excluded.4% to 86.185 Eighty-two percent of women had a BMI more than 30 kg/m² and 33% had BMI more than 40 kg/m². Three babies in the insulin group had birth injuries (one clavicle fracture. The rate of LGA was 3.4 mmol/litre or less.8%) and eight women in the acarbose group (42. The two groups were similar in terms of baseline characteristics and diabetes risk factors but women who were successfully treated using glibenclamide had significantly lower values on the OGTT.93).Diabetes in pregnancy versus 13%). of which one woman discontinued the therapy. 25% in the glibenclamide group and 10.05). admission to NICU (6% versus 7%) or fetal anomalies (2% and 2%).7. this therapy was replaced by insulin therapy. Four women in the glibenclamide group and 41 women in the insulin group had blood glucose concentrations less than 2. macrosomia or caesarean section rates.3 mmol/ litre or less. two symptoms of hypoglycaemia). 28 women (12%) switched to insulin: eight for side effects primarily attributed to hypoglycaemia and 14 for poor control (for six the reason was not clear).6 kg/m². If good control was not achieved within 2 weeks of diet women were offered glibenclamide as an alternative to insulin.13 to 0. had lung complications (8% versus 6%). 95% CI 0. Women in the glibenclamide group were more likely to have pre-eclampsia (adjusted OR 2.006). 24 women assigned to glibenclamide therapy and 19 assigned to acarbose therapy.8 mmol/litre or less.7 mmol/litre and mean pregnancy weight gain 11. An additional 11 (5%) discontinued glibenclamide (most for side effects attributed to hypoglycaemia) and never started insulin.7% in the insulin group. mean fasting level on OGTT 5. one brachial plexus injury and one bone injury). 11 (19%) had babies who weighed more than 4000 g. After logistic regression (adjusting for BMI. but their babies were less likely to be admitted to a NICU (adjusted OR 0.2 mmol/litre or less and 2 hour postprandial 6. Cord-serum insulin concentrations were similar in the two groups and glibenclamide was not detected in the cord serum of any baby in the glibenclamide group. 59 (81%.6 kg.

95% CI 1.Gestational diabetes because of labour abnormalities or suspected macrosomia (compared with 14–17% primary caesarean section rate in the general obstetric population). No hypoglycaemic events were identified in women with gestational diabetes treated by diet alone. There were no differences in the mode of birth. A pilot study randomised 14 women to insulin and 16 to metformin. but not with regular human insulin. [EL = 2−] Education A descriptive study191 compared a hospital outpatient-based nursing intervention and traditional office-based care provided by obstetricians using a research model involving three types of 73 .5% for insulin aspart versus 0. [EL = 1+] A crossover study randomised 15 women with gestational diabetes to no insulin.80 The trial period extended from the diagnosis of gestational diabetes (18–28 weeks) to 6 weeks postpartum. [EL = 1+] An RCT allocated 35 women to either insulin lispro or regular human insulin.4% for human insulin). [EL = 2+] A cohort study compared 27 women with gestational diabetes managed with diet to 30 women treated using insulin and 25 women treated using glibenclamide. number of babies with Apgar score less than 7 at 1 minute or maternal HbA1c between the groups.or long-acting insulin. neonatal complications. OR 4. There were no other differences in neonatal outcomes between the three groups. [EL = 2++] An RCT of metformin for the treatment of gestational diabetes (the MIG trial) is due to report soon.187 The women receiving insulin lispro had significantly lower glucose excursions after a test meal and experienced fewer episodes of hypoglycaemia.4. The study suggests that gestational diabetes can be treated using rapid-acting insulin. There were no symptomatic or significant hypoglycaemic episodes in any group. There were no differences in perinatal outcomes.005).4 Glucose response to a test meal was significantly lower in the lispro group.009.85 The 1 hour postprandial blood glucose values were significantly higher in the regular insulin group than in the lispro or control groups.189 The glucose areas under the curve following a test meal were significantly lower with insulin aspart compared with no insulin. The rate of babies with a cranial : thoracic circumference ratio between the 10th and 25th percentile was significantly higher in the group treated using regular insulin compared with the lispro and control groups. malformations. The safety profile for both groups was similar.3 ± 0. Insulin-meal intervals of 15 minutes avoided preprandial hypoglycaemia without increasing 2 hour postprandial hyperglycaemia and so were beneficial compared with 30 minute intervals. P = 0. [EL = 1+] An RCT compared 24 women with gestational diabetes treated using insulin lispro to 24 women with gestational diabetes treated using regular human insulin and 50 women with a normal GCT.188 Insulin dose was raised in six of the 11 pregnant women on rapid-acting insulin and birthweight was higher in the long-acting insulin group (3943 g ± 492 g versus 3079 g ± 722 g.1 ± 0.75 Insulin analogues An RCT randomly allocated 42 women with insulin-requiring gestational diabetes to either insulin lispro or regular insulin.4–13. insulin aspart or regular human insulin. Occurrence of hypoglycaemia during a 6 week period was 25% lower in the lispro group. There was no difference in obstetric or fetal outcomes. [EL = 1++] A further RCT evaluated pregnancy outcomes in 23 women with gestational diabetes who used rapid. [EL = 1+] A prospective paired study190 tested two insulin-meal intervals to evaluate their effects on shortterm glucose fluctuations in women with tightly controlled gestational diabetes.186 The aim of the study was to assess the prevalence of hypoglycaemia by 72 hour continuous glucose monitoring. A small RCT randomised 27 women with gestational diabetes to either insulin aspart or regular human insulin.9). Asymptomatic hypoglycaemic events were identified in 19 out of 30 (63%) insulin-treated women and 7/25 (28%) glibenclamide-treated women (P = 0. There was no significant difference in mean reduction in HbA1c over the trial period (0. There was no difference between groups in glycosylated protein levels or anti-insulin antibodies. Overall asymptomatic hypoglycaemic episodes were recorded in 26/82 (31%) of women with gestational diabetes in comparison with 0/35 non-diabetic controls.

2. [EL = 2–3] Evidence statement A high-quality RCT has shown that diagnosis and intervention for gestational diabetes (as defined by the WHO) improves maternal and fetal outcomes. In women who are obese. Two studies that may inform future NICE guidelines on the management of diabetes in pregnancy are the HAPO study (an international multicentre prospective cohort study designed to establish the level of hyperglycaemia that results in a significant risk of adverse pregnancy outcome) and the MIG trial (an RCT of metformin for treatment of gestational diabetes). 5. 6. The ACHOIS trial demonstrated potential benefit of treatment for (mild) gestational diabetes. some of whom do not require treatment. trauma during birth to themselves and the baby. Cost-effectiveness The effectiveness of screening. Eight publications were identified in the health economics literature in relation to cost-effectiveness of screening and/or treatment for gestational diabetes. with no difference in other maternal or fetal outcomes. diagnosis and treatment (including monitoring) for gestational diabetes was identified by the GDG as a priority for health economic analysis. but there was a significantly greater risk of having a baby with one or more abnormal outcomes in first-time mothers. 5. neonatal hypoglycaemia. In women requiring hypoglycaemic therapy (those in whom ultrasound investigation suggests incipient fetal macrosomia (abdominal circumference above the 70th percentile) at diagnosis). The nurse-led education did not reduce the risk of abnormal outcomes for women or babies. [EL = 3] Outcomes and risks for the woman and baby No specific searches were undertaken to identify the outcomes and risks for women with gestational diabetes and their babies. Insulin lispro reduces postprandial glucose excursions and hypoglycaemic episodes compared with regular insulin. moderate calorie restriction improves glycaemic control without resulting in ketonaemia. women with gestational diabetes treated pharmacologically and women with gestational diabetes experiencing complications. Publications identified in searches for other sections of the guideline show that women with gestational diabetes are at increased risk of having a macrosomic baby. Women with gestational diabetes are at increased risk of having a macrosomic baby. Exercise in conjunction with diet improves blood glucose control over and above the improvement achieved by diet alone.4. The analysis of cost-effectiveness was addressed through joint work with the GDG for the NICE antenatal care guideline update. Neither study has reported in time for inclusion in this guideline. with no difference in other maternal or fetal outcomes. and may reduce the need for insulin. between 79% and 96% of women will achieve blood glucose targets on glibenclamide. Insulin aspart may reduce postprandial excursions in women with gestational diabetes compared with regular insulin. Often the people who would benefit from treatment must be identified from among a larger group of people. further details are provided in Appendix D. The relative prevalence of maternal hypoglycaemia and LGA babies compared with insulin therapy differs between studies.Diabetes in pregnancy variables: input variables (risk factors before pregnancy). induction of labour and caesarean section. moderating variables (conditions that occur during pregnancy) and outcome variables (normal versus abnormal outcomes for women and babies). This is the case with gestational diabetes.1 and 8. Between 82% and 93% of women with gestational diabetes will achieve blood glucose targets on diet alone. None of the published studies answered the specific cost-effectiveness question addressed by the combined GDGs. perinatal death.5.2). neonatal hypoglycaemia. the cost-effectiveness of screening and treatment for gestational diabetes are highly 74 . A diet that is high in carbohydrates of low glycaemic index improves overall glucose control and reduces postprandial glucose excursions. The methods and results from the health economic modelling are summarised here. perinatal death.153 Evidence of clinical effectiveness is not always sufficient for a treatment to be cost-effective. trauma during birth to themselves and the baby. induction of labour and caesarean section (see Sections 4.

75 . There are currently no data available for comparing the clinical effectiveness of metformin to insulin for the treatment of gestational diabetes. The possibility of universal diagnostic testing was also considered. the cost-effectiveness of different treatment options can be considered separately. with each strategy being compared with a strategy of no screening or treatment. Oman.Gestational diabetes interdependent. Under the baseline assumptions of the health economic model. an analysis of the cost-effectiveness of each individual risk factor followed by an OGTT was conducted.1. As a result.678 compared with no screening or treatment. The health economic model considered screening based on risk factors (age. In the absence of this possibility. respectively.192 67% of UK maternity service providers currently screen using a combination of these factors. ethnicity. Qatar. Approximately 20–50% of women will have a positive screening result using these risk factors with proportions varying considerably from one geographical area to another. oral hypoglycaemic agents (glibenclamide and metformin) and insulin. However. A strategy of offering an OGTT to women from high-risk ethnic backgrounds had an ICER of £3. regardless of which form of treatment they were using. Due to data limitations it was not possible to evaluate the cost-effectiveness of using a combination of the individual risk factors considered in the model (see Appendix D). A strategy of offering an OGTT to all women defined by the American Diabetes Association as being at high risk of gestational diabetes (i. two screening stategies were not dominated by other strategies. Iraq. Lebanon or Egypt).739 compared with screening based on ethnicity alone. FPG and a 1 hour 50 g GCT) followed by a diagnostic test (2 hour 75 g OGTT). Kuwait. Syria. According to a 1999 survey. Using age as a risk factor for screening has a high sensitivity (i.e. Pakistan or Bangladesh) – black Caribbean – Middle Eastern (specifically women whose country of family origin is Saudi Arabia.737 and £5. Given that some form of screening followed by treatment as outlined in the ACHOIS trial is costeffective. United Arab Emirates. £12.209. The ICERs for ethnicity. the MIG trial currently in progress should provide data on the effectiveness of metformin compared with insulin. Given that metformin is even cheaper than glibenclamide. From evidence to recommendations Currently an unselected pregnant population will have the risk of gestational diabetes assessed using the risk factors identified in Section 4.936.5 kg or more previous gestational diabetes family history of diabetes (first-degree relative with diabetes) family origin with a high prevalence of diabetes: – South Asian (specifically women whose country of family origin is India. a single decision-analytic health economic model was developed to help the combined GDGs make recommendations in relation to screening and treatment for gestational diabetes (see Appendix D). since glibenclamide is cheaper than insulin. All women diagnosed with gestational diabetes were assumed to undertake self-monitoring of blood glucose. BMI and family history were £6. namely: • • • • • BMI more than 30 kg/m² previous macrosomic baby weighing 4.e. women older than 25 years or BMI above 27 kg/m² or family history of diabetes or high-risk ethnic background) had an ICER of £21. All of these are below the £20. it is straightforward to show that glibenclamide is cost-effective.000 per quality-adjusted life year (QALY) threshold used by NICE as a willingness to pay for cost-effectiveness and so each one could be regarded as being cost-effective. the combined GDGs expressed concern over the number of women that would undergo an OGTT if the American Diabetes Association screening strategy were recommended because a large proportion of women would be tested based on age criteria alone. it will identify the majority of women with gestational diabetes). BMI and family history of gestational diabetes) and/or blood tests (random blood glucose. An RCT has shown that insulin and glibenclamide are equally effective and. However. Jordan. metformin would be cost-effective if its clinical effectiveness were equal to or greater than that of insulin. The treatment alternatives considered in the model were diet.

Qatar. An RCT investigating the effectiveness of metformin as a treatment for gestational diabetes is due to report imminently.nice. diagnosis and treatment for gestational diabetes has been shown to be costeffective. If near-normal blood glucose is not achieved by diet and exercise alone. Women with gestational diabetes should be informed that they are at increased risk of having a macrosomic baby. trauma during birth to themselves and the baby.5 kg or above previous gestational diabetes family history of diabetes (first-degree relative with diabetes) family origin with a high prevalence of diabetes: – South Asian (specifically women whose country of family origin is India. therefore recommended that hypoglycaemic therapy for women with gestational diabetes be tailored to the glycaemic profile of the individual and acceptability to the woman. hypoglycaemic therapy should be considered. Oman. random blood glucose and urine testing) perform only moderately well in terms of diagnostic value. The biochemical tests considered (GCT. Syria. random blood glucose. In order to make an informed decision about screening and testing for gestational diabetes. Clinically effective hypoglycaemic therapy includes oral hypoglycaemic agents (metformin and glibenclamide) and insulin therapy using regular human insulin or rapidacting insulin analogues. induction of labour and caesarean section. Women with any one of these risk factors should be offered testing for gestational diabetes. Recommendations for screening. Screening for gestational diabetes using fasting plasma glucose. They should also be informed about the role of diet. gestational diabetes will respond to changes in diet and exercise • some women (between 10% and 20%) will need oral hypoglycaemic agents or insulin therapy if diet and exercise are not effective in controlling gestational diabetes • if gestational diabetes is not detected and controlled there is a small risk of birth complications such as shoulder dystocia • a diagnosis of gestational diabetes may lead to increased monitoring and interventions during both pregnancy and labour. The GDG has.  This recommendation is taken from ‘Antenatal care: routine care for the healthy pregnant woman’ (NICE clinical guideline 62). or if ultrasound scans suggest fetal macrosomia. Lebanon or Egypt). FPG. Iraq. Jordan. Kuwait. the following risk factors for gestational diabetes should be determined: • • • • • body mass index above 30 kg/m² previous macrosomic baby weighing 4. the importance of maternal glycaemic control during labour and birth and early feeding of the baby in order to reduce the risk of neonatal hypoglycaemia. and the risk of the baby developing obesity and/or diabetes in later life. diagnosis and treatment for gestational diabetes Screening for gestational diabetes using risk factors is recommended in a healthy population. available from www. glucose challenge test and urinalysis for glucose should not be undertaken.Diabetes in pregnancy Whilst screening using risk factors is less sensitive than performing a GCT or OGTT. perinatal death. it is more practical and less disruptive for women.uk/CG062. women should be informed that: • in most women. the GDG for the diabetes in pregnancy guideline is of the opinion that gestational diabetes should be treated initially with diet and exercise. body weight and exercise. If treatment with metformin is shown to be as effective as treatment with insulin then it is likely to be costeffective because metformin is considerably cheaper than glibenclamide.org. At the booking appointment. Pakistan or Bangladesh) – black Caribbean – Middle Eastern (specifically women whose country of family origin is Saudi Arabia. United Arab Emirates. Health economic analysis has demonstrated that glibenclamide is costeffective. neonatal hypoglycaemia. Given that screening. 76 . but the clinical evidence comes from a healthcare setting outside the UK and the GDG’s view is that the acceptability to women of treating gestational diabetes with glibenclamide has not been demonstrated in the NHS healthcare setting.

Women with any of the other risk factors for gestational diabetes should be offered an OGTT at 24–28 weeks. which have not been systematically evaluated for clinical and cost-effectiveness (including acceptability) within the UK. rapid-acting insulin analogues [aspart and lispro] and/or hypoglycaemic agents [metformin and glibenclamide] should be tailored to the glycaemic profile of.  Fasting plasma venous glucose concentration greater than or equal to 7. induction of labour and caesarean section • the importance of maternal glycaemic control during labour and birth and early feeding of the baby in order to reduce the risk of neonatal hypoglycaemia • the possibility of transient morbidity in the baby during the neonatal period. trauma during birth (to themselves and the baby). carbohydrates from low glycaemic index sources. Hypoglycaemic therapy should be considered for women with gestational diabetes if ultrasound investigation suggests incipient fetal macrosomia (abdominal circumference above the 70th percentile) at diagnosis. Hypoglycaemic therapy for women with gestational diabetes (which may include regular insulin. diabetes should not be treated with metformin but insulin should be used to maintain blood glucose levels. There is strong evidence for its effectiveness and safety. neonatal hypoglycaemia and perinatal death. Informed consent on the use of glibenclamide in pregnancy should be obtained and documented. diagnosis and treatment for gestational diabetes What is the clinical and cost-effectiveness of the three main available screening techniques for gestational diabetes: risk factors.  Glibenclamide is used in UK clinical practice in the management of diabetes in pregnancy and lactation.0 mmol/litre or 2 hour plasma venous glucose concentration greater than or equal to 7. where possible. and universal OGTT (with or without fasting)? Why this is important Following the Australian carbohydrate intolerance study in pregnant women (ACHOIS) it seems that systematic screening for gestational diabetes may be beneficial to the UK population. Research recommendations for screening. Report of a WHO consultation. 77 . This evidence is not currently reflected in the SPC (July 2008). A multicentre randomised controlled trial is required to test the existing screening techniques. Women with gestational diabetes whose pre-pregnancy body mass index was above 27 kg/m² should be advised to restrict calorie intake (to 25 kcal/kg/day or less) and to take moderate exercise (of at least 30 minutes daily). Women with gestational diabetes should be informed that good glycaemic control throughout pregnancy will reduce the risk of fetal macrosomia.8 mmol/litre.  Metformin is used in UK clinical practice in the management of diabetes in pregnancy and lactation. the individual woman. body weight and exercise • the increased risk of having a baby who is large for gestational age. Women who have had gestational diabetes in a previous pregnancy should be offered early self-monitoring of blood glucose or OGTT at 16–18 weeks. This evidence is not currently reflected in the SPC (July 2008). which increases the likelihood of birth trauma. induction of labour or caesarean section. Women with gestational diabetes should be instructed in self-monitoring of blood glucose. and a further OGTT at 28 weeks if the results are normal. Part 1: diagnosis and classification of diabetes mellitus. Women with gestational diabetes should be offered information covering: • the role of diet. Women with gestational diabetes should be advised to choose.Gestational diabetes The 2 hour 75 g oral glucose tolerance test (OGTT) should be used to test for gestational diabetes and diagnosis made using the criteria defined by the World Health Organization. There is strong evidence for its effectiveness and safety. Hypoglycaemic therapy should be considered for women with gestational diabetes if diet and exercise fail to maintain blood glucose targets during a period of 1–2 weeks. Geneva: World Health Organization. The SPC advises that glibenclamide is contraindicated in pregnancy. World Health Organization Department of Noncommunicable Disease Surveillance (1999) Definition. two-stage screening by the glucose challenge test and OGTT. which may require admission to the neonatal unit • the risk of the baby developing obesity and/or diabetes in later life. lean proteins including oily fish and a balance of polyunsaturated fats and monounsaturated fats. Targets for blood glucose control should be determined in the same way as for women with pre-existing diabetes. diagnosis and classification of diabetes mellitus and its complications. and acceptability to. Informed consent on the use of metformin in these situations should be obtained and documented. The SPC advises that when a patient plans to become pregnant and during pregnancy.

01).1%.5 5.3% and also had birthweight macrosomia. augmentation of labour (28% versus 15%.3%.1 Antenatal care Target ranges for blood glucose during pregnancy Description of the evidence Fetal macrosomia The term macrosomia is often used to describe birthweight more than 4000 g.6 to 2.01).0 mmol/litre during the second and third trimesters were significantly more likely to have a macrosomic baby (28% versus 17%. ten had HbA1c more than 6.01).0 mmol/litre during the second and third trimesters (n = 43) to women with mean CBG more than 6.03).0 mmol/litre (n = 32). Women in the conventional group were significantly more likely to have a macrosomic baby (13. P < 0.6% versus 7. P < 0.05). obstetric history or family history.196 Of these 13. RR 2. In symmetric macrosomia the baby is big but the only potential problem is trauma during birth. obesity. P < 0. Symmetric macrosomia accounts for about 70% of cases. The baby is at risk of shoulder dystocia. In the conventional management group women were instructed by a nurse and were assessed weekly for fasting and 3 hour postprandial venous plasma glucose during clinic visits. induction of labour (27% versus 22%. P < 0.193 [EL = 3] Eleven studies were identified that examined the relationship between birthweight and blood glucose control during pregnancy. Symmetric macrosomia may also occur in women without diabetes. Women in the intensive group had a mean blood glucose of 5 ± 2 mmol/litre. There are two types of macrosomia: symmetric and asymmetric. lunch and dinner). P < 0. polyhydramnios (amniotic fluid index 25 cm or more or one fluid pocket 8 cm or more) or fat line (5 mm or more subcutaneous tissue near the level of the umbilical vein insertion).0001). [EL = 2++] In a cohort study of 66 women with diabetes (22 with type 1 diabetes. Asymmetric macrosomia is characterised by thoracic and abdominal circumference that is relatively larger than the head circumference. as a consequence. 41 with gestational diabetes) 13 (20%) had at least one of the following ultrasound markers for macrosomia: estimated fetal weight > 90th percentile for gestational age using the Williams growth curve. clavicular fracture and brachial palsy and.7 days.3 days versus 3. 2 hour postprandial and at bedtime). In the intensive management group women were assigned memory reflectance meters and were instructed by nurses in self-monitoring of blood glucose (seven times a day: fasting. P < 0. four with type 2 diabetes.07. [EL = 2++] 78 . It is also sometimes used to refer to birthweight ≥ 90th percentile for gestational age. The women performed four daily self-monitored blood glucose determinations with glucose strips (fasting and 2 hours after breakfast.01) and a baby admitted to intensive care (25% versus 6. hypertension and type 2 diabetes later in life. coronary heart disease. Women with mean CBG more than 6. ethnicity. a longer hospital stay (mean 4. the woman is more likely to undergo caesarean section. There were no significant differences between groups with regard to parity. preprandial. P < 0. It has also been suggested that babies with asymmetric macrosomia may be at increased risk of obesity.0001). which is also referred to as LGA. A non-randomised intervention study in women with gestational diabetes194 compared intensive (n = 1316) and conventional (n = 1145) management.2 mmol/litre or more. caesarean section (19% versus 13%. [EL = 2++] A cohort study of 75 women with type 1 diabetes195 compared women with mean capillary blood glucose (CBG) less than 6. All women were diagnosed by a 1 hour glucose challenge test (GCT) followed by a 100 g OGTT if plasma glucose was 7. There was a significant difference in HbA1c between those with markers and those without (P < 0.8. 95% CI 1. Assignment to groups depended on the availability of reflectance meters.

P < 0.5 ± 0. smoking status. P < 0. but it explained less than 5% of the variance. a greater success in achieving glycaemic control targets (55% versus 30%. [EL = 2++] A cohort study compared 323 women with type 1 diabetes and 361 women without diabetes. maternal weight gain and glucose control during the first two trimesters did not influence the incidence of LGA. The HbA1 at birth was significantly higher in women whose babies who were LGA than those who had appropriate-for-gestational-age (AGA) babies (8. There were also fewer instances of shoulder dystocia (3% versus 18%) and third or fourth degree perineal laceration (9% versus 24%). of degrees of hyperglycaemia and overgrowth. The pre-pregnancy weight.05).5% versus 13. weight gain during pregnancy. P < 0.05). P = 0. number of prior miscarriages.05).8%) despite good glycaemic control. weight gain in pregnancy and adherence to therapy were similar in the two groups.3–5.8 mmol/ litre and postprandial level not exceeding 7.200 Women with diabetes were significantly more likely to have an LGA baby (> 90th percentile) than women without diabetes (18.001).2%. social class.4 ± 0. even in the normal range.05). a mean blood glucose level of 4. A possible explanation for the finding is that HbA1c is not predictive of macrosomia as it fails to detect the more relevant postprandial peaks in blood glucose. The findings from this study indicate that there is a continuum. [EL = 2++] An RCT involving 66 women with gestational diabetes155 who required insulin therapy at 30 weeks of gestation or earlier randomly assigned women to monitor either preprandial or 1 hour postprandial blood glucose levels.001).01).05). The strongest predictor of birthweight was third-trimester non-fasting glucose levels (P = 0.7 mmol/litre). as measured by HbA1c. [EL = 2+] HbA1c or postprandial peaks A cohort study of 289 women with type 1 diabetes199 found high rates of macrosomia (48.3.Antenatal care A cohort study of 127 women with type 1 diabetes197 found 43% of babies were LGA.001).3 mmol/ litre were associated with an increased risk of SGA babies (18%) compared with values above this level (1%).84 mmol/litre. Women in the postprandial group were significantly less likely to have a caesarean section for cephalopelvic disproportion (12% versus 36%.1 ± 1. gestational age at diagnosis of gestational diabetes and at birth. The postprandial monitoring group had a significantly reduced incidence of pre-eclampsia (3% versus 21%. P = 0. [EL = 1+] A study of 51 pregnant women without diabetes203 who monitored their blood glucose from 28–38 weeks found overall mean blood glucose during the third trimester to be 3. P = 0. [EL = 2++] A cohort study compared 52 women with type 1 diabetes to 52 women without diabetes. Maternal age. 6 ± 0. [EL = 2++] A cohort study of 111 women with type 1 diabetes201 found macrosomia in 32 (29%) babies.4–4.3% versus 7. parity. The mean peak postprandial response occurred 1 hour postprandially and never exceeded 5. There was a significant association between fetal abdominal circumference and 1 hour postprandial glucose values (P < 0. HbA1c in the third trimester was the most powerful predictor of macrosomia. degree of adherence to therapy and degree of achievement of target blood glucose concentrations were similar in both groups.155 The goal of insulin therapy was a preprandial value of 3.9 mmol/litre or a postprandial value less than 7.11 mmol/ litre. P < 0.1%.198 The women with diabetes maintained ‘normal’ blood glucose levels from the 12 weeks of gestation (defined as a fasting level of 3.0003) and a significant negative relationship between head : abdominal circumference ratio (P = 0.6 mmol/litre.04) and a baby with neonatal hypoglycaemia (3% versus 21%. There were 3/33 (9%) macrosomic babies in the postprandial monitoring group compared with 12/33 (36%) in the preprandial monitoring group (P = 0. In logistic regression the only variable that was associated with macrosomia was postprandial glucose between 29 and 32 weeks of gestation (P < 0. Insulin dose. age at onset of diabetes.001) and smaller neonatal triceps skinfold thickness (4.204 79 .0–3. [EL = 1+] An RCT of 61 women with type 1 diabetes202 randomly assigned women at 16 weeks of gestation to preprandial or postprandial blood glucose monitoring. [EL = 2++] Respiratory distress In a study of 180 babies in 179 women (20 with gestational diabetes and 160 with type 1 diabetes) mean blood glucose was calculated for the period from gestational week 30–32 until birth.8 mmol/litre. Postprandial values less than 7.9 versus 5. The mean infant body weight was 2910 g with no babies above the 75th percentile.

45 women.0 mmol/litre (n = 32).2% to 17%) in the lowest quintile of HbA1c measurements to 79% (95% CI 60% to 91%) in the highest quintile.6 mmol/litre. [EL = 2+] A cohort study of 75 women with type 1 diabetes195 compared women with mean CBG less than 6.3%. from 12% (95% CI 7.198 The women with diabetes maintained ‘normal’ blood glucose levels (defined as a fasting level of 3.Diabetes in pregnancy Respiratory distress was significantly more frequent in women with mean blood glucose more than 5. 16 and 20 weeks of gestation.91–137.7 mmol/litre. There were no cases of respiratory distress in babies of women with type 1 diabetes compared to five cases in women without diabetes (P < 0. stillbirth or neonatal death had worse glycaemic control before and during pregnancy. morbidity of the baby diagnosed within the first month of life and mortality). Wales and Northern Ireland with pre-existing type 1 or type 2 diabetes who gave birth between March 2002 and February 2003 (3808 pregnancies) reported that women who had babies with malformations. In one study the aim was to keep all blood glucose levels less than 5.96 95% CI 4. There were no detected differences in perinatal outcome.7%).204 Malformations and perinatal death were more frequent in women with mean blood glucose more than 5.4 mmol/litre and postprandial values less than 6.01). Respiratory distress was significantly more frequent in babies of women with mean CBG above 6.26). [EL = 2++] Tight versus very tight control A systematic review and meta-analysis of tight versus very tight control for diabetes in pregnancy identified two small RCTs involving 182 women. OR 25. but not at 8 or 24 weeks of gestation.8 mmol/litre and postprandial level not exceeding 7. One hundred and sixty-five (29%) of the pregnancies terminated with adverse outcomes (congenital anomaly. a mean blood glucose level of 4.5 mmol/litre.1. [EL = 2+] Preterm labour A cohort study of 145 women with type 1 diabetes recruited during the first trimester considered the relationship between glycaemic control and preterm labour. P < 0. HbA1c levels above 7% showed almost linear association with risk of adverse outcome. [EL = 2++] Congenital malformations.0 mmol/litre (21.0–3.8% versus 2.0 mmol/litre (n = 43) to women with mean capillary glucose more than 6. There were significant differences in HbA1c between the group with minor malformations and the group without malformations at 12.4–4. The prevalence of adverse outcomes varied six-fold. [EL = 2++] A cohort study compared 52 women with type 1 diabetes to 52 women without diabetes. stillbirth and neonatal death A cohort study of women with type 1 diabetes considered the relationship between glycaemic control during pregnancy and the incidence of minor congenital malformations. The data show a dose-dependent association between HbA1c levels above 7% in the first trimester and risk of adverse pregnancy outcome.206 Thirty-nine minor congenital malformations were diagnosed in 32 babies (18.03). 80 . [EL = 2+] In a cohort study of 180 babies in 179 women (20 with gestational diabetes and 160 with type 1 diabetes) mean blood glucose was calculated for the period from gestational week 30–32 until birth.7 mmol/litre) from 12 weeks of gestation.205 Logistic regression found HbA1c at 26 weeks of gestation and urogenital infection to predict the onset of preterm labour.208 [EL = 1+] Current practice A CEMACH audit2 of 3733 women in England.5% increased risk of adverse outcome. In the other study the aim for the very tight group were fasting values less than 4. with every 1% increase in HbA1c corresponding to 5.5 mmol/litre (P < 0. Maternal hypoglycaemia was common among women whose diabetic control was very tight (1 trial. [EL = 3] The proportion of women with HbA1c less than 7% in each trimester is shown in Table 5. [EL = 2+] The association between first-trimester HbA1c levels and risk of adverse pregnancy outcomes of women with type 1 diabetes was evaluated in a cohort study207 involving 573 pregnant women with type 1 diabetes.6 mmol/litre in the very tight group.05).

[EL = 3–4] The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that 65% of women with type 1 diabetes and 54% of women with type 2 diabetes had sub-optimal glycaemic control in the first trimester (P = 0. The enquiry panels identified that there were incidents of failure to change insulin regimens.33 Of the women with poor pregnancy outcome 71% (146/205) were documented as having suboptimal control after the first trimester compared with 37% (77/209) of the women with good pregnancy outcome (OR 5. Of the women with poor pregnancy outcome 84% (171/204) were documented as having sub-optimal control in the first trimester compared with 61% (117/192) of the women with good pregnancy outcome (OR 3. Postprandial blood glucose levels have a stronger association with incidence of macrosomia than HbA1c. 95% CI 3.9 mmol/litre and 1 hour postprandial blood glucose below 7.9% Normally formed stillbirths or 12. the postprandial blood glucose target for women during pregnancy should be lower than when not pregnant.9% neonatal deaths Normally formed and alive at 28.0% < 13 weeks of gestation 25.33 [EL = 3–4] Existing guidance The NSF for diabetes20 recommends regular blood glucose monitoring with the aim of maintaining blood glucose within the normal non-diabetic range.0% 66. women with diabetes should aim to keep fasting blood glucose between 3.7). Evidence statement There is evidence that high blood glucose levels during pregnancy are associated with fetal macrosomia (and associated birth injury and caesarean sections) and fetal morbidity.8).8 mmol/litre during pregnancy.3% 18–23 weeks of gestation 57. meaning that any apparent reduction 81 . 95% CI 2.7% 19.2. No evidence was identified to support the clinical utility of HbA1c measurements in the second and third trimesters of pregnancy. therefore.5% 28 days The CEMACH enquiry found 38% of the women who had an HbA1c measurement had a recorded a value of less that 7%.4. OR 2. From evidence to recommendations The evidence suggests that the important intervention for preventing fetal macrosomia is blunting postprandial blood glucose peaks and.0 95% CI 1.7% 70.Antenatal care Table 5. the targets for blood glucose control during pregnancy should be agreed with the individual woman. There is evidence that in women without diabetes blood glucose levels in the third trimester of pregnancy are lower than in non-pregnant women.4% 40.1 to 5. The GDG’s view is that HbA1c is not a reliable indicator of glycaemic control in the second and third trimesters of pregnancy because of physiological changes that occur in all pregnant women and lead to reduced HbA1c in women without diabetes.9% 27 weeks of gestation 58% 40.1 to 3.2 The CEMACH case–control study found sub-optimal glycaemic control was associated with poor pregnancy outcome.5 and 5. taking into account the risk of hypoglycaemia. If it is safely achievable.1 Proportion of women with HbA1c less than 7% in each trimester and birth outcomes2 Pre-pregnancy Malformations 22. As in the preconception period.3 to 8.2). non-responsive local strategy of diabetes antenatal care and problems within the diabetes multidisciplinary team in both groups of women.3% 39. No local targets were set for glycaemic control during pregnancy for more women who had a poor pregnancy outcome than for women who had a good pregnancy outcome (49% [44/90] versus 33% [28/86].064). After the first trimester 41% of women with type 1 diabetes and 35% of women with type 2 diabetes had sub-optimal glycaemic control. Two RCTs found that monitoring of postprandial blood glucose produced better outcomes than preprandial monitoring.

5 mmol/litre or less. Recommendations for target ranges for blood glucose during pregnancy Individualised targets for self-monitoring of blood glucose should be agreed with women with diabetes in pregnancy. P = 0. P = 0. taking into account the risk of hypoglycaemia. preprandial levels 5. All blood glucose values as well as insulin doses and dietary intake were recorded. There were also fewer instances of shoulder dystocia (3% versus 18%) and third.001) and smaller neonatal triceps skinfold thickness (4. Women in the postprandial group were significantly less likely to have a caesarean section for cephalopelvic disproportion (12% versus 36%.5 and 5. The postprandial monitoring group had a significantly reduced incidence of pre-eclampsia (3% versus 21%. P < 0.05). The postprandial protocol required daily monitoring of blood glucose concentrations before breakfast (fasting) and 1 hour after each meal.1 ± 1.04) or a baby with neonatal hypoglycaemia (3% versus 21%.155 The ethnic background of the sample was 85% Hispanic.2 Monitoring blood glucose and ketones during pregnancy Description of the evidence Two RCTs were identified that investigated preprandial versus postprandial monitoring of blood glucose during pregnancy. not be used routinely for assessing glycaemic control in the second and third trimesters of pregnancy. CBG readings were measured by using a memory-based glucose reflectance meter.5 mmol/litre and 5. 11% white and 4% black or Asian.202 All women were on a four-times-daily basal bolus insulin regimen. There were 3/33 (9%) macrosomic babies in the postprandial monitoring group compared with 12/33 (36%) in the preprandial monitoring group (P = 0.9 mmol/litre and 1 hour postprandial blood glucose below 7. Insulin doses and glucose readings were also recorded by diary and brought to the clinic.01). The women measured their blood glucose concentration using memory-based reflectance glucometers.5 ± 0. [EL = 1++] The ACHOIS trial randomly assigned 1000 women with gestational diabetes to either an intervention group or routine care. therefore. The first study consisted of 61 women with type 1 diabetes who were randomly assigned at 16 weeks of gestation to either preprandial or postprandial blood glucose monitoring. P < 0. women with diabetes should aim to keep fasting blood glucose between 3. The rate of serious perinatal outcomes among babies 82 . There were no research recommendations relating to the target ranges for blood glucose during pregnancy. The preprandial group was asked to monitor before breakfast and preprandially.Diabetes in pregnancy in HbA1c in women with diabetes during the second and third trimesters of pregnancy does not necessarily indicate improved glycaemic control. This represents a change in clinical practice to avoid unnecessary and misleading HbA1c measurements.5 mmol/litre or less and 2 hour postprandial levels 7. P = 0. Women were randomly assigned to monitor either preprandial or 1 hour postprandial blood glucose levels. [EL = 1++] The second study consisted of 66 women with gestational diabetes who required insulin therapy.0 mmol/litre or less) followed by daily monitoring at rotating times. The preprandial monitoring protocol required daily monitoring of fasting. HbA1c should. preprandial and bedtime CBG concentrations. HbA1c should not be used routinely for assessing glycaemic control in the second and third trimesters of pregnancy.153 The intervention was a package of care that included instructions on self-monitoring of blood glucose four times daily until blood glucose levels had been in the recommended range for 2 weeks (fasting glucose levels more than 3.3. greater success in achieving glycaemic control targets (55% versus 30%. 5. The postprandial group was asked to monitor before breakfast and 1 hour after meals.8 mmol/litre during pregnancy.or fourth-degree perineal laceration (9% versus 24%).9 versus 5. The package of care also included insulin therapy with the dose adjusted on the basis of glucose levels and individualised dietary advice from a qualified dietitian. If it is safely achievable.05).05).

Antenatal care

was significantly lower in the intervention group (1% versus 4%, P = 0.01). The number needed to treat to prevent a serious outcome in a baby was 34. There was no significant difference between groups in maternal quality of life. [EL = 1++] Three studies were identified that reported on the use of continuous blood glucose monitoring in women with diabetes. Two cohort studies were in women with type 1 diabetes209,210 [EL = 2+] and one case series was in women with gestational diabetes.211 [EL = 3] All three studies reported hyperglycaemic episodes undetected by self-monitoring of blood glucose. These episodes were usually due to the consumption of high carbohydrate food between meals and were undetected by self-monitoring protocols that required testing only after main meals. The three studies showed that examining 72 hour glucose profiles can help to identify patterns of glucose control, better target insulin treatment, assist in patient education and improve dietary adherence. A retrospective study120 examined the effect of an intensive diabetes management programme during pregnancy on women’s long-term self-management behaviours and glycaemic control. There was a significant improvement in all diabetes self-management behaviours, including frequency of self-monitoring of blood glucose, frequency of insulin injections, and frequency and complexity of insulin dose adjustment from entry to the programme to the baby’s birth. There was also a significant improvement in HbA1c from entry to the baby’s birth. [EL = 2−] An RCT212 investigated whether glycaemic control achieved by women using telephone modems for the transmission of self-monitored blood glucose data was better than that achieved by women managed in a similar fashion without modem connection. The study showed that telemedicine is a practical way of providing specialist care to pregnant women. [EL = 1+] A systematic review of observational studies213 investigated the risk of adverse pregnancy outcomes in pregnant women with diabetes in relation to glycaemic control. The review showed that an increase in adverse pregnancy outcomes in women with diabetes who had poor glycaemic control (congenital malformations, pooled OR 3.44, 95% CI 2.30 to 5.15; risk reduction of congenital malformation 0.39–0.59 for each 1% decrease in HbA1c; miscarriage, pooled OR 3.23, 95% CI 1.64 to 6.36; perinatal mortality, pooled OR 3.03, 95% CI 1.87 to 4.92). [EL = 3] No studies were identified that assessed how ketones should be monitored during pregnancy.

Existing guidance
The NSF for diabetes20 recommends that ‘women should be supported and encouraged to monitor their blood glucose regularly’.

Evidence statement
Two high quality RCTs have found better pregnancy outcomes for women with diabetes when blood glucose is monitored 1 hour after meals than when it is monitored before meals. One RCT found that a treatment package that included self-monitoring of blood glucose improved outcomes in women with gestational diabetes compared with routine obstetric care. Two cohort studies and a case series showed that self-monitoring of blood glucose undertaken only after main meals may not detect hyperglycaemia following the consumption of food between meals. No studies were found on monitoring for ketones during pregnancy.

From evidence to recommendations
The evidence regarding the effectiveness of self-monitoring of blood glucose 1 hour after meals for improving pregnancy outcomes suggests that postprandial monitoring should not be restricted to main meals. The effectiveness of monitoring using meters supports the provision of such meters (see Section 3.5). The GDG’s view is that women with insulin-treated diabetes are vulnerable to nocturnal hypoglycaemia during pregnancy and that it is good clinical practice to undertake an additional test before going to bed at night. As no evidence was identified in relation to monitoring for ketones during pregnancy, the GDG’s consensus recommendation based on current best practice is to provide urine and blood ketone testing strips and to advise women to test ketone levels if they are hyperglycaemic or unwell. 83

Diabetes in pregnancy

Recommendations for monitoring blood glucose and ketones during pregnancy
Women with diabetes should be advised to test fasting blood glucose levels and blood glucose levels 1 hour after every meal during pregnancy. Women with insulin-treated diabetes should be advised to test blood glucose levels before going to bed at night during pregnancy. Women with type 1 diabetes who are pregnant should be offered ketone testing strips and advised to test for ketonuria or ketonaemia if they become hyperglycaemic or unwell.

Research recommendations for monitoring blood glucose and ketones during pregnancy
How effective is ambulatory continuous blood glucose monitoring in pregnancies complicated by diabetes? Why this is important The technology for performing ambulatory continuous blood glucose monitoring is only just becoming available, so there is currently no evidence to assess its effectiveness outside the laboratory situation. Research is needed to determine whether the technology is likely to have a place in the clinical management of diabetes in pregnancy. The new technology may identify women in whom short-term postprandial peaks of glycaemia are not detected by intermittent blood glucose testing. The aim of monitoring is to adjust insulin regimens to reduce the incidence of adverse outcomes of pregnancy (for example, fetal macrosomia, caesarean section and neonatal hypoglycaemia), so these outcomes should be assessed as part of the research.

5.3

Management of diabetes during pregnancy
Description of the evidence
Hypoglycaemia Hypoglycaemia significantly affects maternal quality of life and increases the risk of physical injury. During pregnancy the frequency of hypoglycaemia may increase due to intensification of treatment, an impairment of counter-regulatory hormonal responses and an increased risk of hypoglycaemia unawareness.214,215 Pregnancy nausea and vomiting can also contribute to hypoglycaemia due to fluctuations in carbohydrate ingestion. [EL = 2+]
A cohort study of 84 pregnant women with type 1 diabetes undergoing intensified treatment68 found hypoglycaemia requiring third-party assistance occurred in 71% of women, with a peak incidence between 10–15 weeks of gestation. Consequences of maternal hypoglycaemia included several grand mal seizures, five episodes of cerebral oedema, and two road traffic accidents and comminuted fracture of the tibia and fibula. [EL = 2++]

Hyperemesis gravidarum Severe nausea and vomiting in pregnant women with diabetes can lead to ketoacidosis, and DKA during pregnancy carries a risk of fetal death.
Two case studies were identified that reported healthy live births to women with diabetes and hyperemesis gravidarum following treatment with parenteral nutrition.216,217 One case study reported a fetal death in a woman with hyperemesis gravidarum and DKA following a delay in treatment.218 [EL = 3]

Diabetic ketoacidosis A case series of 37 women admitted with DKA219 concluded that vomiting and the use of betamimetic drugs were the primary cause in 57% of cases. Non-adherence to treatment and physician management errors were the primary cause in 24% of cases and contributory in 16%.

84

Antenatal care

Common physician management errors included the use of urine instead of blood to monitor maternal glucose control, failure to adhere to pregnancy standards of glucose control and failure to employ home blood glucose monitoring. [EL = 3] A cohort study of 257 people220 with DKA admitted to a large urban teaching hospital compared outcomes in people treated by a general physician (n = 224) with those in people treated by a physician with subspecialty training in diabetes (n = 33). People treated by a diabetes specialist had shorter length of stay (3.3 versus 4.9 days, P < 0.0043) and incurred lower hospital charges ($5,463 versus $10,109, P < 0.0001). Plasma glucose in generalist-treated people took longer to fall to less than 11.1 mmol/litre and they had a higher rate of readmission for DKA than the specialist-treated people (6% versus 2%, P = 0.03). [EL = 2+]

Rapid-acting insulin analogues Rapid-acting insulin analogues (aspart and lispro) confer the following benefits compared with regular insulin outside pregnancy:221
• • • • fewer episodes of hypoglycaemia a reduction in postprandial glucose excursions an improvement in overall glycaemic control an improvement in patient satisfaction.

These benefits have also been demonstrated in the pregnant population (see Section 3.6).

Long-acting insulin analogues The NICE guideline for the management of type 1 diabetes recommends the long-acting insulin analogue glargine for use outside of pregnancy.221 However, no clinical trials have as yet been published for their use in pregnancy (see Section 3.6). Four-times-daily versus twice-daily insulin regimens An open label RCT compared glycaemic control and perinatal outcomes in pregnant women with diabetes using two different insulin regimens.222 One hundred and thirty-eight women with gestational diabetes and 58 with pre-existing diabetes received insulin four times daily, and 136 women with gestational diabetes and 60 with pre-existing diabetes received insulin twice daily. Glycaemic control was better with the four-times-daily regimen than with the twice-daily regimen. In women with gestational diabetes the four-times-daily regimen resulted in a lower rate of overall neonatal morbidity than the twice-daily regimen. Four-times-daily rather than twicedaily insulin improved glycaemic control and perinatal outcomes without increasing the risks of maternal hypoglycaemia and caesarean section. [EL = 1++] Continuous subcutaneous insulin infusion (insulin pump therapy) The most widespread method of administering insulin is via subcutaneous insulin injections using a basal/bolus regimen consisting of a basal dose of long-acting insulin, usually administered with a pen before bed, and bolus of rapid-acting insulin given before meals. This is often referred to as a multiple daily injection (MDI) regimen. Insulin can also be administered using CSII (also known as insulin pump therapy). Both regular insulin and rapid-acting insulin can be administered by pump. The potential benefits of CSII are reduced risk of hypoglycaemia, decreased risk of fasting hyperglycaemia and improved adherence as the woman does not have to constantly inject insulin.223
NICE guidance for the non-pregnant population concluded that, compared with MDI regimens, CSII results in a modest but worthwhile improvement in blood glucose control and quality of life.14 A systematic review224 investigated the effectiveness of insulin delivery via CSII as compared with MDI regimens for the treatment of diabetes during pregnancy in women with pre-existing diabetes or gestational diabetes. Only two studies were included in the review, neither included women with gestational diabetes. There was a significant increase in mean birthweight associated with CSII as opposed to MDI (two trials, 61 participants, weighted mean difference (WMD) 220.56, 95% CI −2.09 to 443.20). However, taking into consideration the lack of significant difference in rate of macrosomia (birthweight greater than 4000 g; RR 3.20, 95% CI 0.14 to 72.62), this finding was not viewed by the authors as being clinically significant. There were no significant differences in perinatal outcomes between CSII and MDI (perinatal mortality, including stillbirths from 24 weeks of gestation and neonatal deaths up to 7 days of life, RR 2.00, 95% CI 0.20 to 19.91; 85

Diabetes in pregnancy

fetal anomaly, RR 1.07, 95% CI 0.07 to 15.54; gestational age at birth, WMD 0.63, 95% CI −4.87 to 6.13; neonatal hypoglycaemia, RR 1.00, 95% CI 0.07 to 14.64; and SGA, RR 1.55, 95% CI 0.27 to 9.00). Neither were there any significant differences in maternal outcomes between CSII and MDI (caesarean section rate, RR 1.03, 95% CI 0.57 to 1.84; mean maternal HbA1c; 24 hour mean blood glucose level in each trimester; hypoglycaemia; or hyperglycaemia). [EL1+] Three further RCTs of CSII in women with type 1 diabetes during pregnancy were identified.225–227 The studies included a total of 200 women. There were no significant differences between groups in glycaemic control or in obstetric or neonatal outcomes. There were four cases of ketoacidosis in women using pumps. This was attributed to catheter occlusion (one case), catheter leakage (one case) and pump failure (one case). Another case was reported but without attribution. [EL = 1++] One RCT was designed to assess the effect of CSII on retinopathy.226 This study of 40 women with type 1 diabetes reported progression to proliferative retinopathy in two women using CSII. This was attributed to rapid and significant improvement in glycaemic control. [EL = 1+] Three cohort studies compared outcomes in women using CSII with those in women using MDI regimens. [EL = 2+]228 [EL = 2+]229 [EL = 2++]230 In each study women were offered pump therapy due to difficulties achieving glycaemic control. All studies reported good glycaemic control and obstetric and neonatal outcomes.

Current practice The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that women with type 1 diabetes were more likely to experience recurrent episodes of hypoglycaemia than women with type 2 diabetes (P < 0.001), with 61% (105/171) of women with type 1 diabetes and 21% (25/121) of the women with type 2 diabetes having recurrent episodes of hypoglycaemia.33 One or more episodes of hypoglycaemia required help in 25% (33/133) of the women with type 1 diabetes and 4% (4/102) of the women with type 2 diabetes. [EL = 3–4]

Existing guidance
The NICE technology appraisal relating to insulin pump therapy (CSII) for people with type 1 diabetes states that insulin pumps can be used in pregnancy even if there is good glycaemic control on MDI regimens.14 The Driver and Vehicle Licensing Agency (DVLA) medical rules for drivers do not include any special considerations for diabetes in pregnancy. Fitness to drive is assessed on the basis of the risk of hypoglycaemia, regardless of whether or not the driver is pregnant (see www.dvla.gov.uk/medical. aspx and www.direct.gov.uk/en/Motoring/DriverLicensing/MedicalRulesForDrivers/index.htm).

Evidence statement
During pregnancy women with diabetes treated using insulin are at an increased risk of hypoglycaemia and hypoglycaemia unawareness. Rapid-acting insulin analogues (aspart and lispro) are associated with fewer episodes of hypoglycaemia compared with regular human insulin. When compared with regular human insulin the use of rapid-acting insulin analogues during pregnancy has also been associated with a reduction in postprandial glucose excursions, an improvement in overall glycaemic control and an improvement in patient satisfaction. Ketoacidosis is a complication that can result in fetal death. Outcomes may be improved with prompt assessment and treatment by a health professional with specialist diabetes training. RCTs have shown similar outcomes in women using CSII and MDI regimens. Ketoacidosis may result from pump failure. Cohort studies have reported good outcomes in women offered pump therapy because of difficulty achieving glycaemic control using MDI regimens.

From evidence to recommendations
Women and their partners should be informed of the increased risk of hypoglycaemia and hypoglycaemia unawareness during pregnancy, and information about prevention, recognition and treatment (including the provision of a concentrated glucose solution and, if they have type 1 86

Antenatal care

diabetes, glucagon, and education in their use) should be reinforced in women with insulintreated diabetes who are pregnant. Women with insulin-treated diabetes should also be advised of the consequences of hypoglycaemia and the dangers associated with driving during periods of hypoglycaemia unawareness. They should be encouraged to carry something that identifies them as having diabetes so they can be treated promptly if disabling hypoglycaemia occurs. The evidence supports the use of the rapid-acting insulin analogues aspart and lispro in women with diabetes in pregnancy, and also insulin pump therapy (CSII) in women who have difficulty achieving glycaemic control without disabling hypoglycaemia. Since DKA can be accelerated in pregnancy and is associated with maternal and fetal adverse outcomes (including fetal death), the GDG’s consensus view is that current best practice should be followed, i.e. DKA should be excluded in women with type 1 diabetes who become unwell in pregnancy and pregnant women with DKA should be admitted immediately for level 2 critical care where they can receive medical and obstetric care.

Recommendations for management of diabetes during pregnancy
Healthcare professionals should be aware that the rapid-acting insulin analogues (aspart and lispro) have advantages over soluble human insulin during pregnancy and should consider their use. Women with insulin-treated diabetes should be advised of the risks of hypoglycaemia and hypoglycaemia unawareness in pregnancy, particularly in the first trimester. During pregnancy, women with insulin-treated diabetes should be provided with a concentrated glucose solution and women with type 1 diabetes should also be given glucagon; women and their partners or other family members should be instructed in their use. During pregnancy, women with insulin-treated diabetes should be offered continuous subcutaneous insulin infusion (CSII or insulin pump therapy) if adequate glycaemic control is not obtained by multiple daily injections of insulin without significant disabling hypoglycaemia. During pregnancy, women with type 1 diabetes who become unwell should have diabetic ketoacidosis excluded as a matter of urgency. During pregnancy, women who are suspected of having diabetic ketoacidosis should be admitted immediately for level 2 critical care, where they can receive both medical and obstetric care.

Research recommendations for management of diabetes during pregnancy
Do new-generation CSII pumps offer an advantage over traditional intermittent insulin injections in terms of pregnancy outcomes in women with type 1 diabetes? Why this is important Randomised controlled trials have shown no advantage or disadvantage of using CSII pumps over intermittent insulin injections in pregnancy. A new generation of CSII pumps may offer technological advantages that would make a randomised controlled trial appropriate, particularly with the availability of insulin analogues (which may have improved the effectiveness of intermittent insulin injections). 

Level 2 critical care is defined as care for patients requiring detailed observation or intervention, including support for a single failing organ system or postoperative care and those ‘stepping down’ from higher levels of care.  For the purpose of this guidance, ‘disabling hypoglycaemia’ means the repeated and unpredicted occurrence of hypoglycaemia requiring third-party assistance that results in continuing anxiety about recurrence and is associated with significant adverse effect on quality of life.

87

Diabetes in pregnancy

5.4

Retinal assessment during pregnancy
There are two widely used classifications for grading the levels of diabetic retinopathy. The English classification for sight-threatening diabetic retinopathy is used for two-field diabetic retinopathy screening (see the National Screening Committee’s diabetic retinopathy screening programme for England and Wales, available at www.retinalscreening.nhs.uk/). The English classification progresses from no diabetic retinopathy, to background diabetic retinopathy, to pre-proliferative diabetic retinopathy, to proliferative diabetic retinopathy (PDR). The Early Treatment Diabetic Retinopathy Study (ETDRS) classification has provided an evidence base for progression of diabetic retinopathy based on grading of seven-field stereo-photographs. The ETDRS classification progresses from no diabetic retinopathy, to mild non-proliferative diabetic retinopathy (NPDR), to moderate NPDR, to moderately severe NPDR, to severe NPDR, to PDR. The lesions found within the two classifications follow a common language: background diabetic retinopathy is equivalent to mild NPDR and is characterised by increased vascular permeability; pre-proliferative diabetic retinopathy is equivalent to moderate NPDR, moderately severe NPDR and severe NPDR; in PDR the development of new blood vessels can significantly reduce vision. The progression from no diabetic retinopathy to PDR normally occurs over a period of years, but sudden worsening may occur in pregnancy. Duration of diabetes is known to be an important factor in the progression of diabetic retinopathy and in the development of PDR in people with diabetes. Data from an epidemiological study showed that PDR varied from 1.2% to 67% in people who had had diabetes for less than 10 years and more than 35 years, respectively.231

Description of the evidence
A cohort study232 determined the prevalence of retinopathy characteristically seen in people with diabetes and IGT, and in people with new-onset diabetes of known duration in the Diabetes Prevention Program (DPP) cohort. The DPP recruited and followed people with elevated fasting glucose (5.3–6.9 mmol/litre) and IGT with no history of diabetes other than gestational diabetes that did not persist after pregnancy. A random sample of 302 participants who developed diabetes and those who remained free from diabetes after 3 years follow-up was used for the retinopathy study. Retinopathy consistent with diabetic retinopathy was detected in 12.6% of people with diabetes and 7.9% of people without diabetes (P = 0.03). The study suggests that retinopathy is present in people with elevated fasting glucose and IGT with no known history of diabetes. [EL = 2+]

Progression of diabetic retinopathy Three cohort studies were identified that found pregnancy to be independently associated with progression of diabetic retinopathy.
One study compared 180 women who became pregnant during an RCT of intensive versus conventional treatment of type 1 diabetes with women who did not become pregnant.233 In the intensive treatment group 693/2950 (23%) non-pregnant women had progression of retinopathy compared with 39/124 (31%) pregnant women (OR 1.62, 95% CI 1.01 to 2.59, P < 0.05). In the conventional treatment group 1742/5605 (31%) non-pregnant women had progression of retinopathy compared with 37/73 (50.3%) pregnant women (OR 2.54, 95% CI 1.59 to 4.03, P < 0.0001). [EL = 2++] A cohort study compared 60 pregnant women with type 1 diabetes to 80 non-pregnant women with type 1 diabetes.234 Progression of retinopathy occurred in 10/35 women with pre-existing retinopathy in the pregnant group. There was no progression of retinopathy in the non-pregnant controls (24 had pre-existing retinopathy). [EL = 2+] A cohort study compared 171 pregnant women with type 1 diabetes to 298 non-pregnant women with type 1 diabetes.235 A multivariate analysis (pregnancy, HbA1c, blood pressure, number of previous pregnancies and duration of diabetes) found pregnancy to be independently associated with progression of retinopathy (OR 1.8, 95% CI 1.1 to 2.8, P < 0.02). [EL = 2++]

Severity of retinopathy at conception Four studies found progression of retinopathy during pregnancy to be associated with severity of retinopathy at conception.
88

PDR developed in 2/32 (6%) women with mild NPDR and 9/31 (29%) women with moderate NPDR at conception.5%) progressed to PDR necessitating photocoagulation. and 5/8 (63%) women with PDR in the first trimester progressed (P = 0. but that duration of diabetes may be an important factor in the development of PDR. [EL = 2+] A cohort study followed 154 women with type 1 diabetes. Magnitude of improvement in glycaemic control: Three studies233.236 [EL = 2++] Progression of retinopathy following commencement of intensive treatment has also been observed in non-pregnant adults with diabetes. 8/38 (21. during each trimester and 12 months postpartum.236 The study found progression of two steps or more in 4/39 (10.001).007) and in women with moderate to severe NPDR at booking (6/163 versus 3/10.242 All studies found poor glycaemic control to be associated with progression of retinopathy during pregnancy. [EL = 2++] Duration of diabetes Six cohort studies of women with type 1 diabetes found progression of retinopathy during pregnancy was associated with duration of diabetes. P = 0. 12 (55%) had NPDR progression and 5 (22.0001).237 Diabetic retinopathy deteriorated during pregnancy in all five women with proliferative retinopathy at baseline.01).5%) showed no progression. 5 (22.238 The study found 38 women had no retinopathy at conception. The difference in progression of retinopathy between these two groups was statistically significant (P = 0. [EL = 2++] Glycaemic control Seven studies considered the effect of glycaemic control on the progression of diabetic retinopathy during pregnancy.238–241 The effect of duration of diabetes on progression of retinopathy during pregnancy is difficult to separate from the effect of the severity of retinopathy at conception as the two are correlated.01). Twenty-two women had NPDR at conception. However.233. The study included 20 pregnancies in women with duration of diabetes more than 20 years who had no or mild retinopathy at booking and of these only one progressed.1%) women with microaneurysms only.Antenatal care A cohort study of 155 pregnant women with type 1 diabetes found women with more severe retinopathy at conception were more likely to show progression during pregnancy (χ² for trend.3%) women with no retinopathy.239 Twenty-three percent (18/78) of women with no retinopathy in the first trimester progressed.243 involving 1441 people with 89 .8%) women with mild NPDR and 17/31 (54. [EL = 2++] A recent cohort study of 179 pregnancies in 139 women with type 1 diabetes241 found progression of retinopathy was significantly increased in women with duration of diabetes of 10– 19 years compared with duration less than 10 years (8/80 versus 0/71. [EL = 2++] A cohort study of 35 women with type 1 diabetes found progression during pregnancy in 3/10 women with no retinopathy at baseline and in 3/20 with background retinopathy (2/20 developed proliferative retinopathy). This suggests that severity of retinopathy at conception is more important than duration of diabetes for the progression of diabetes during pregnancy. P = 0. 28/68 (41%) women with NPDR in the first trimester progressed. retinopathy progressed by two steps or more in 55% of women with 15 years or less of duration of diabetes and 50% of women with more than 15 years of duration of diabetes. A study of 155 pregnant women with type 1 diabetes236 found that in women with moderate or more severe retinopathy at baseline.237. 5/32 (18. The effect of large improvement of glycaemic control is difficult to separate from poor glycaemic control as women with the largest improvement were those who had poor initial control.239 found that a large improvement in glycaemic control in the first trimester was associated with progression of retinopathy. 28 (74%) showed no progression and ten (26%) progressed to mild NPDR. Women with no retinopathy or only microaneurysms at conception did not develop PDR.8%) women with moderate NPDR.236. [EL = 2+] A cohort study evaluated 65 women with type 1 diabetes before pregnancy.243–246 In the DCCT study70.234. P < 0.235–239. PDR developed in only 18% of women with 15 years or less of duration of diabetes compared with 39% of women with more than 15 years of duration of diabetes. This suggests that severity of retinopathy at conception may be more important than duration of diabetes in the progression of retinopathy during pregnancy.

Focal treatment also increased the chance of visual improvement. Thirteen women experienced postpartum regression.005). focal photocoagulation was effective in reducing the risk of moderate visual loss but that scatter photocoagulation was not. the results showed that 9.3%. [EL = 2+] Laser treatment for diabetic macular oedema A large multicentre RCT247 in which people with macular oedema and mild or moderate diabetic retinopathy in one or both eyes were randomly assigned to focal argon laser photocoagulation (754 eyes) or deferred photocoagulation (1490 eyes) showed that focal photocoagulation substantially reduced the risk of visual loss (12% versus 24% at 3 year follow-up).02) to be associated with progression of retinopathy. None of the women who developed PDR during pregnancy experienced postpartum regression. Ten showed mild progression during pregnancy. [EL = 2++] A cohort study of 65 women with type 1 diabetes were followed until 12 months postpartum.1% of people in the intensive group compared with 7. It was found that for eyes with macular oedema. Improvement in the macular oedema occurred in 64.2% had a deterioration in visual acuity equivalent to a doubling of the visual angle and 3.251 was conducted in 546 people undergoing their first photocoagulation treatment for maculopathy.001). Overall people with early worsening of retinopathy in the intensive group had a 74% reduction in the risk of subsequent progression as compared with people with early worsening who received conventional treatment (P < 0.243 [EL = 1++] Hypertension A cohort study of 154 women with type 1 diabetes examined the effect of hypertension on the progression of retinopathy during pregnancy. However this RCT did not mention whether pregnant women were included.239 Multiple regression found pregnancy-induced hypertension (P = 0. 69% in the intensive group and 57% in the conventional group had shown complete recovery by the 18 month visit.01) and chronic hypertension (P = 0.6% and exudates in 77. At 9 month follow-up. five progressed to PDR. It defined the concepts of ‘clinically significant macular oedema’ and ‘treatable lesions’. [EL = 1+] An audit by the UK National Diabetic Retinopathy Laser Treatment group250. [EL = 2++] Postpartum regression A cohort study followed 154 women with type 1 diabetes through pregnancy to 12 weeks postpartum. [EL = 1++] Logistic regression incorporating both initial HbA1c and the change between initial HbA1c and 4 month HbA1c found the latter to be the dominant factor for early worsening of retinopathy.001). twelve progressed from mild to severe NPDR. [EL = 2++] A cohort study of 65 pregnant women with type 1 diabetes238 found systolic blood pressure to be higher in women who showed progression of retinopathy during pregnancy than in those who did not (P < 0. of which two showed regression postpartum. There was no development of PDR in the group with no retinopathy at conception. [EL = 3–4] 90 .Diabetes in pregnancy type 1 diabetes (726 with no retinopathy at baseline) progression of retinopathy was observed at the 6 and/or 12 month visit in 13. of which five showed complete postpartum regression.3% of eyes had a visual acuity less than 6/60. There was no evidence that people with more rapid reduction of HbA1c had a greater risk of early worsening of retinopathy than people with more gradual reduction when the reductions were of similar magnitude. Twenty-two women had NPDR at conception: five of these women experienced no change during pregnancy.239 Fifty-one women had progression of retinopathy during pregnancy of which seven developed PDR. decreased the frequency of persistent macular oedema and caused only minor visual field losses. [EL = 3–4] A subsequent RCT249 compared eyes selected for early photocoagulation in the first RCT247 by treating with one of four combinations of scatter (panretinal) and focal treatment. Among people who had experienced early worsening of retinopathy. Of these 28 showed no change during pregnancy.6% in the conventional group (P < 0. [EL = 1+] A further report248 from the RCT described treatment techniques in detail.238 Thirty-eight women had no retinopathy at conception.

even in the absence of preretinal or vitreous haemorrhage. At 9 month follow-up neovascularisation had regressed fully in 50.240 [EL = 3] The Diabetic Retinopathy Study group252 recommend treatment for control eyes with ‘high risk characteristics’.0001). P = 0.004) of deterioration in visual acuity by three lines of the ETDRS chart. hyperglycaemia (higher HbA1c) and with not smoking. male sex.0004) and a 47% reduced risk (99% CI 7% to 70%. The presence of high-normal blood pressure (> 90th percentile but less than 141/90 mm Hg) resulted in a prospectively higher occurrence of retinopathy and of progression of pre-existing retinopathy. They reported four retinopathy factors that increase the 2 year risk of developing severe visual loss: presence of vitreous or preretinal haemorrhage. This audit showed that regression of neovascularisation was associated with greater areas of retinal ablation at the initial treatment session. The 2 year risk of severe visual loss without treatment outweighed the risk of harmful treatment effects for eyes with new vessels and preretinal or vitreous haemorrhage and for eyes with new vessels on or within one disc diameter of the optic disc (NVD) equalling or exceeding one-quarter to one-third of the disc area. [EL 3–4] A cross-sectional study255 in Norway of 600 people with a mean age of 19.0001) and mean arterial blood pressure (P = 0.3 years follow-up no difference was found in the incidence between the intensive and moderate groups with regard to the progression of diabetic retinopathy. [EL = 1+] The UK National Diabetic Retinopathy Laser Treatment audit251 was conducted on 546 people undergoing their first photocoagulation treatment for maculopathy.0001) were significantly associated with retinopathy. progression was associated with older age. presence of new vessels. [EL 2+] An RCT257 investigated the effect of tight blood pressure control and risk of microvascular complications in people with type 2 diabetes. glycaemic exposure over 6 years.009 ). an increase in the HbA1c level and an increase in diastolic blood pressure level from the baseline to the 4 year follow-up. location of new vessels on or near the optic disc and severity of new vessels. [EL 2++] A cross-sectional study256 in the USA of 634 people with type 1 diabetes diagnosed before age 30 years evaluated retinopathy after 14 years. [EL = 3–4] An RCT253 that compared photocoagulation with no treatment found that photocoagulation reduced the risk of severe visual loss by 50% or more. [EL = 3–4] Effect of blood pressure on macular oedema and diabetic retinopathy A study254 investigated the relationship between blood pressure and diabetic retinopathy in 249 young people with type 1 diabetes. whereas the increased risk of a person developing macular oedema was associated with the presence of gross proteinuria at baseline.Antenatal care Laser treatment for proliferatvie diabetic retinopathy A study of 55 pregnant women with type 1 diabetes found that progression of retinal disease was arrested with photocoagulation during pregnancy in four women with proliferative retinopathy. At 5. [EL 1+] An RCT260 compared tight blood pressure control (blood pressure less than 150/85) with less tight blood pressure control (blood pressure less than 180/105) and its relationship with 91 .8% of cases with proliferative retinopathy. higher blood pressure and with not smoking. [EL 2++] A prospective RCT259 compared the effects of intensive and moderate blood pressure control on the incidence and progression of type 2 diabetic complications in 470 people. duration of diabetes (P = 0. and there was no change or deterioration in 10. P = 0. age (P = 0. Development of retinopathy (incidence) was strongly associated with baseline glycaemia. Progression was more likely with higher HbA1c or diastolic blood pressure at baseline. Retinopathy was present in 63% of young people and hypertension in 2%.3%. The increased risk of proliferative retinopathy was associated with the presence of hypertension at baseline. In those who already had retinopathy. In a multiple logistic regression model. higher mean HbA1c (P = 0.8 years evaluated the association of various risk factors with retinopathy. [EL 1++] A cross-sectional study258 investigated risk factors related to the incidence and progression of diabetic retinopathy from diagnosis over 6 years in 1919 people with type 2 diabetes. After 9 years follow-up the group assigned to tight blood pressure control had a 34% reduction in risk in the proportion of participants with deterioration of retinopathy by two steps (99% CI 11% to 50%.

Diabetes in pregnancy

diabetic retinopathy in 1148 people with diabetes. At 4.5 years follow-up people allocated to tight blood pressure control were less likely to undergo photocoagulation (RR 0.65, P = 0.03). This difference was driven by a difference in photocoagulation due to maculopathy (RR 0.58, P = 0.02). [EL = 1+]

Current practice The CEMACH enquiry reported that a detailed retinal assessment was recorded in the woman’s notes at least once during pregnancy in 79.9% of women with pre-existing diabetes.2 The CEMACH case– control study reported that women with poor pregnancy outcome were as likely not to have a retinal assessment during the first trimester or at booking if later (36% [70/194]) than women who had a good pregnancy outcome (27% [49/183], OR 1.4, 95% CI 0.9 to 2.2, adjusted for maternal age and deprivation).33 Only 55% of the 258 assessments were recorded to have been done through dilated pupils and for 40% of women details about the retinal assessment procedure were not documented. The most common concern noted by the CEMACH enquiry panels over sub-optimal diabetes care in pregnancy was sub-optimal retinal function monitoring and management. [EL = 3–4]
The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that women with type 1 diabetes were more likely to have retinopathy than women with type 2 diabetes (P < 0.001), with 36% (50/138) of women with type 1 diabetes and 9% (9/96) of the women with type 2 diabetes having retinopathy in pregnancy.33 This was a new finding in 26% (13/50) of the women with type 1 diabetes and 56% (5/9) of the women with type 2 diabetes. Of the women with pre-existing retinopathy there was evidence of deterioration in 18% of the women with type 1 diabetes and 11% of the women with type 2 diabetes. Women with type 1 diabetes were more likely to have a retinal assessment compared to women with type 2 diabetes (78% versus 64%, P = 0.02). Where retinopathy was found both groups of women were as likely to be referred to an ophthalmologist (35% versus 44%, P = 0.62). [EL = 3–4]

Existing guidance
The NSF for diabetes recommends full retinal assessment in all women with pre-existing diabetes during the first trimester (or at booking if this is later).20

Evidence statement
In some women pregnancy may accelerate progression of diabetic retinopathy. This is more likely in women with more severe diabetic retinopathy, poor glycaemic control and hypertension. Some diabetic retinopathy may regress spontaneously after the woman has given birth. It is difficult to separate the influence of different factors which have been found to be associated with progression of retinopathy during pregnancy. The magnitude of improvement in glycaemic control is associated with glycaemic control prior to conception (which in turn is associated with duration of diabetes and severity of diabetes at conception). The DCCT found the magnitude, but not the rapidity, of the reduction in HbA1c during the first 6 months of intensive treatment to be an important risk factor for early worsening of diabetic retinopathy. Whether or not the risk of retinopathy progression can be reduced by more gradual reduction in glycaemic control can be resolved only by an RCT. Evidence supports the use of laser treatment in diabetic macular oedema. Further evidence shows that control of blood pressure has a positive effect on macular oedema and progression of diabetic retinopathy.

From evidence to recommendations
Given the evidence of a rapid change in diabetic retinopathy during pregnancy (because of persistent hyperglycaemia) the GDG’s view is that healthcare professionals should err on the side of caution by offering increased frequency of surveillance in the preconception period and throughout pregnancy to women with long-standing poor glycaemic control and pre-proliferative diabetic retinopathy or PDR, and by treating pre-existing diabetic retinopathy before conception. There is evidence that rapid optimisation of glycaemic control can worsen diabetic retinopathy. However, it is the GDG’s view that the benefits to the fetus of good glycaemic control outweigh

92

Antenatal care

the risks to the woman (early worsening of diabetic retinopathy). Healthcare professionals should, therefore, encourage improvement in glycaemic control in pregnancy and address ophthalmological complications of diabetes during pregnancy if they occur. In most women, laser treatment can be performed during pregnancy and will reduce the risks of sight loss as a result of progression of diabetic retinopathy. Careful control of blood pressure will also have a positive effect on sight-threatening diabetic retinopathy. Only in very rare circumstances might early birth be considered to reduce the risks of vision loss in pregnancy. The GDG’s view is that retinal assessment during pregnancy for women with diabetes should be conducted in accordance with the recommendations of the National Screening Committee’s diabetic retinopathy screening programme (that is, retinal assessment should be performed using digital imaging with mydriasis (dilation of the pupils) using tropicamide). In making its recommendations, the GDG has also noted the report of the CEMACH diabetes in pregnancy audit, which highlighted that only 55% of women with pre-existing diabetes were documented to have received retinal assessment through dilated pupils. If diabetic retinopathy is found to be present in early pregnancy, referral should usually be guided by the standard referral criteria (see the National Screening Committee’s website), and women should be seen by an ophthalmologist within 4 weeks, except for PDR when urgent referral is required. As with preconception care, if there are concerns in relation to the possible worsening of diabetic retinopathy with imminent improvement of very poor blood glucose control then referral with lesser degrees of retinopathy may be considered. The recommendations in relation to retinal assessment in the preconception period are presented in Section 3.10.

Recommendations for retinal assessment during pregnancy
Pregnant women with pre-existing diabetes should be offered retinal assessment by digital imaging with mydriasis using tropicamide following their first antenatal clinic appointment and again at 28 weeks if the first assessment is normal. If any diabetic retinopathy is present, an additional retinal assessment should be performed at 16–20 weeks. If retinal assessment has not been performed in the preceding 12 months, it should be offered as soon as possible after the first contact in pregnancy in women with pre-existing diabetes. Diabetic retinopathy should not be considered a contraindication to rapid optimisation of glycaemic control in women who present with a high HbA1c in early pregnancy. Women who have preproliferative diabetic retinopathy diagnosed during pregnancy should have ophthalmological follow-up for at least 6 months following the birth of the baby. Diabetic retinopathy should not be considered a contraindication to vaginal birth.

Research recommendations for retinal assessment during pregnancy
Should retinal assessment during pregnancy be offered to women diagnosed with gestational diabetes who are suspected of having pre-existing diabetes? Why this is important Women with gestational diabetes may have previously unrecognised type 2 diabetes with retinopathy. At present this is not screened for because of the difficulty in identifying these women amongst the larger group who have reversible gestational diabetes. The benefit of recognising such women is that treatment for diabetic retinopathy is available and could prevent short- or long-term deterioration of visual acuity. The research needed would be an observational study of retinal assessment in women newly diagnosed with gestational diabetes to determine whether there is a significant amount of retinal disease present. The severity of any abnormality detected might identify women most at risk for appropriate retinal assessment.

93

Diabetes in pregnancy

5.5

Renal assessment during pregnancy
Diabetic nephropathy is a progressive disease that can be divided into the following stages:261 [EL = 4] • microalbuminuria (incipient nephropathy) – small amounts of albumin are excreted in the urine • macroalbuminuria or proteinuria (overt nephropathy) – widespread glomerular sclerosis resulting in progressively larger amounts of protein excreted in the urine • end-stage renal disease – decreasing creatinine clearance, increasing serum creatinine and uraemia.

Description of the evidence
Effect of pregnancy on progression of nephropathy A systematic review262 considered the effects of pregnancy on diabetic nephropathy. The review included 11 longitudinal studies involving a total of 201 people. Only one study had a nonpregnant control group. The other studies compared the average rate of decline in renal function with the expected rate of decline in the general non-pregnant population of people with diabetic nephropathy. The review found that most studies suggest that pregnancy is not associated with development of nephropathy or with accelerated progression of pre-existing nephropathy, with the exception of women with moderate to advanced disease where pregnancy may accelerate progression to end-stage renal disease. [EL = 2++] Effect of nephropathy on pregnancy outcome A systematic review262 which included 11 studies and 681 people found that women with diabetic nephropathy were at increased risk of adverse pregnancy outcomes, in particular IUGR, chronic hypertension, pre-eclampsia and preterm birth (see Table 5.2). Pre-eclampsia and preterm birth were associated with incipient nephropathy (microalbuminuria) as well as overt nephropathy. [EL = 1++]
Table 5.2
Outcome Chronic hypertension Pre-eclampsia Caesarean section Intrauterine growth restriction Birth before 34 weeks

Outcome of pregnancy in women with diabetic nephropathy262
Occurrence (%) 23–77 15–64 63–86 9–45 16–45

A cohort study was identified that had been published since the systematic review. The cohort study considered pregnancy outcome in women with type 1 diabetes and microalbuminuria.263 Of 240 consecutive pregnancies, 203 women (85%) had normal urinary albumin excretion, 26 (11%) had microalbuminuria and 11 (5%) had diabetic nephropathy. In this study normal urinary albumin excretion was defined as less than 30 mg/24 hours, microalbuminuria was defined as urinary albumin excretion 30–300 mg/24 hours and diabetic nephropathy was defined as urinary albumin excretion more than 300 mg/24 hours. The incidence of pre-eclampsia was 6% in women with normal urinary albumin excretion, 42% in women with microalbuminuria and 64% in women with nephropathy (P < 0.001). The incidence of preterm birth (before 34 weeks) was 6% in women with normal urinary albumin excretion, 23% in women with microalbuminuria and 45% in women with diabetic nephropathy (P < 0.001). The incidence of SGA babies was 2% in women with normal urinary albumin excretion, 4% in women with microalbuminuria and 45% in women with diabetic nephropathy (P < 0.001). [EL = 2++]

Antihypertensive treatment for microalbuminuria A cohort study involving 46 women evaluated the impact of antihypertensive treatment with methyldopa in normotensive pregnant women with type 1 diabetes and microalbuminuria.264 The women were similar in terms of age, diabetes duration, pre-pregnancy BMI, HbA1c and blood
94

Antenatal care

pressure, and all were referred before 17 weeks of gestation. The prevalence of preterm birth before 34 weeks of gestation was reduced from 23% to 0% (P = 0.02); the prevalence of preterm birth before 37 weeks of gestation was reduced from 62% to 40% (P = 0.15); and the prevalence of pre-eclampsia was reduced from 42% to 20% (P = 0.11). Perinatal mortality occurred in 4% versus 0%. [EL = 2++]

Current practice The CEMACH enquiry reported that women who had a poor pregnancy outcome were more likely not to have monitoring for nephropathy (22% [46/209]) than women who had a good pregnancy outcome (13% [26/206], OR 1.9, 95% CI 1.1 to 3.3, adjusted for maternal age and deprivation). In an additional case–control analysis lack of monitoring for nephropathy was associated only with fetal congenital anomaly and not with fetal or neonatal death after 20 weeks of gestation; it is therefore unlikely to have been causative for poor pregnancy outcome. Nephropathy itself was not associated with poor pregnancy outcome. One of the most common concerns noted by the CEMACH enquiry panels over sub-optimal diabetes care in pregnancy was sub-optimal renal function monitoring and management.33 [EL = 3–4]
The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that there was no significant difference in the rate on nephropathy in pregnancy in women with type 1 or type 2 diabetes, with 8% (12/148) of women with type 1 diabetes and 5% (6/119) of the women with type 2 diabetes having nephropathy during their pregnancy.33 Women with type 1 diabetes were as likely to have monitoring for nephropathy as women with type 2 diabetes (86% versus 79%, P = 0.60). Where nephropathy was found both groups of women were as likely to have a test of renal function (75% versus 50%, P = 0.29). [EL = 3–4] The CEMACH enquiry did not state an explicit standard of monitoring for nephropathy, but it recommended that appropriate monitoring included testing for microalbuminuria (incipient nephropathy) via protein dipstick testing of urine or serum creatinine.33 [EL = 3–4]

Existing guidance
The NICE guideline for type 2 diabetes defines microalbuminuria as albumin : creatinine ratio 3.5 mg/mmol or more (for women) or albumin concentration 20 mg/litre or more. Macroalbuminuria is defined as albumin : creatinine ratio 30 mg/mmol or more or albumin concentration 200 mg/litre or more.8 The NICE guidelines for type 1 and type 2 diabetes in adults recommend annual testing for nephropathy using urine albumin : creatinine ratio and serum creatinine. It is recommended that people with nephropathy have measurements of urine albumin and serum creatinine levels at each visit.7,8

Evidence statement
In the majority of studies pregnancy has not been associated with the development of nephropathy or with accelerated progression of pre-existing nephropathy. Data from three studies suggest that in women with moderate to advanced disease pregnancy may accelerate progression to endstage renal disease. All stages of nephropathy, including microalbuminuria, are associated with adverse pregnancy outcomes, especially IUGR, pre-eclampsia and preterm birth. A small cohort study suggested that antihypertensive treatment with methyldopa in women with type 1 diabetes and microalbuminuria reduced the risk of preterm birth (before 34 weeks of gestation). No evidence was identified in relation to thromboprophylaxis in the presence of macroalbuminuria.

From evidence to recommendations
NICE recommends that renal assessment outside pregnancy should use urine albumin : creatinine ratio and serum creatinine. Estimated glomerular filtration rate (eGFR) should not be used during 95

Diabetes in pregnancy

pregnancy as it underestimates the glomerular filtration rate.439 There is no evidence on the optimal assessment schedule during pregnancy. As both microalbuminuria and macroalbuminuria are associated with adverse outcomes the GDG recommends assessment in the preconception period or at the first presentation after conception. All pregnant women should have their urine tested for proteinuria as part of routine antenatal care (see the NICE antenatal care guideline).9 If serum creatinine is abnormal (120 micromol/litre or more) or if total protein excretion exceeds 2 g/day, referral to a nephrologist should be considered. No evidence was identified in relation to thromboprophylaxis in the presence of macroalbuminuria. The GDG’s consensus view is that healthcare professionals should follow best current practice in terms of thromboprophylaxis for women with diabetes and macroalbuminuria (antenatal administration of aspirin for proteinuria less than 5 mg/day and heparin for proteinuria more than 5 mg/day; planned early birth may need to be considered because of the risk of developing pre-eclampsia). The recommendations in relation to renal assessment in the preconception period are presented in Section 3.11.

Recommendations for renal assessment during pregnancy
If renal assessment has not been undertaken in the preceding 12 months in women with preexisting diabetes, it should be arranged at the first contact in pregnancy. If serum creatinine is abnormal (120 micromol/litre or more) or if total protein excretion exceeds 2 g/day, referral to a nephrologist should be considered (eGFR should not be used during pregnancy). Thromboprophylaxis should be considered for women with proteinuria above 5 g/day (macroalbuminuria).

Research recommendations for renal assessment during pregnancy
Does identification of microalbuminuria during pregnancy offer the opportunity for appropriate pharmacological treatment to prevent progression to pre-eclampsia in women with pre-existing diabetes? Why this is important Microalbuminuria testing is available, but it is not performed routinely in antenatal clinics for women with pre-existing diabetes because a place for prophylactic treatment of pre-eclampsia in microalbuminuria-positive women has not been investigated. The benefit of clinically and cost-effective prophylactic treatment would be to significantly improve pregnancy outcomes in this group of women.

5.6

Screening for congenital malformations
Description of the evidence
Women with diabetes have an increased risk of having a baby with a congenital malformation. Major congenital malformations affecting babies of women with diabetes include cardiac, neural tube and genitourinary anomalies. Table 5.3 lists anomalies associated with diabetes as well as the estimated prevalence and RR compared to women without diabetes, as reported in published studies.265 More recent data from the CEMACH enquiry found the prevalence of confirmed major anomalies to be 41.8 per 1000 total births (live and stillborn).2 Separate rates for babies of women with type 1 diabetes (n = 1707) and type 2 diabetes (n = 652) born between 1 March 2002 and 28 February 2003 are summarised in Table 5.4.266 Women with type 2 diabetes were more likely to come from a Black, Asian or Other Minority Ethnic group (type 1 diabetes 9.1%, type 2 diabetes 48.8%). Perinatal mortality in babies of women with diabetes was 31.8 per 1000 births, nearly four times higher than the general maternity population. One hundred and ninety-seven major congenital anomalies were confirmed in 148 babies. The prevalence of major congenital anomaly

96

0 3–5 Caudal regression syndrome Central nervous system 0.8 (0.3 to 3.5 (0. observed (expected)a (expected)a 81 (37) 6 (2.1) – – – One or more malformations Neural tube defects Other central nervous system Eye Ear Congenital heart disease Cleft lip and palate Cleft palate Digestive system External genital system Limb Trisomy 21 Other chromosomal a Internal urogenital system 9 (6.4) 0 (0.5) 15 (10.5 to 4.9) 0 (0.5 5–2 Table 5.1 Microcephaly Isolated hydrocephlus Duodenal atresia Anorectal atresia Hypoplastic left colon Talipes Arthrogryposis Uretal duplication Cystic kidney Renal dysgenesis Hydronephrosis 1.5 Neural tube defects (including anencephaly) 2.6) 0 (0.7) 4 0 2 Standardised prevalence ratio for babies of women with both type1 and type 2 diabetes (95% CI) 2.6 to 2.1 to 7.0 200 2–10 Gastrointestinal 3 Musculoskeletal system Orofacial cleft Urinary tract 0.2) 1.9) 0 (1.9) 0 (0.8) 1.4 1.8 to 2.5 to 3.2 (0.9) 1 (2.8–2.8) 4 (0.9) 0.3) 2 (0.4 Observed and expected prevalence of congenital malformations in babies of women with type 1 and type 2 diabetes (from CEMACH)266 Type of malformation Babies of women with Babies of women with type 1 diabetes. observed type 2 diabetes.267 97 .4 (2.Antenatal care Table 5.0 to 7.0 (0.5 (0.2–0.4) 1.8 to 2.6) 3 (2.3) 2 (0.2 (1.9) 4 (3.2 to 2.0) – 3.4) 0 (0.7–3.4 (0.3 Detectable major congenital malformations in babies of women with pre-existing diabetes265 Group of malformations Cardiac Specific malformations Transposition of the great arteries Ventricular septal defect Coarctation of the aorta Atrial septal defect Asymmetric septal hypertrophy Prevalence per Relative risk compared to 100 births women without diabetes 3.5) 1.6) 4.2) 2 2 28 (12.1) Other (non-chromosomal) 6 Expected rates are based on European Surveillance of Congenital Anomalies (EUROCAT) 2002.3) 9(3.6) – 1.7) 33 (8.0–10.2 (2.6) 1.5) 0 (0.2 to 5.0) 1 (2.3) 2 (1.7) 1 (2.8 1.4) 5 (1.0 2–20 1.6 (0.

269 The study compared 79 women with type 1 diabetes to 16 366 women without diabetes. however published studies have shown that some biochemical markers tend to be lower in women with type 1 diabetes than in women without diabetes. Anomalies other than congenital heart disease were diagnosed antenatally in 71. According to the NICE antenatal care guideline. and ensuring appropriate obstetric management. 253 women with type 1 diabetes). Women should understand that it is their choice to embark on screening for Down’s syndrome. women without diabetes MoM 1. Information about screening options for Down’s syndrome that can be understood by all women. A meta-analysis268 included published studies of differences in AFP (14 studies. Screening should be performed by the end of the first trimester (14 weeks of gestation). 43 per 1000 births for type 2 diabetes). 687 women with type 1 diabetes). total hCG (nine studies. 95% CI 0. P = 0.16.96 for total hCG. clinical practice has been to make adjustments when calculating the risk of anomalies for women with type 1 diabetes to take account of these differences. Congenital heart disease was diagnosed antenatally in 54.94 for uE3. This section considers what additional screening should be offered to women with diabetes. The weight-corrected median multiple-of-median (MoM) was 0.270 Levels of PAPP-A were significantly lower in women with type 1 diabetes (P = 0. free β-hCG (one study. alpha fetoprotein (AFP). Women with diabetes do not have an increased risk of chromosomal anomalies.00.024). unconjugated estriol (uE3) and inhibin A). 0. The increase was mainly due to an increase in neural tube defects (4. including those whose first language is not English. beta human chorionic gonadotropin [β-hCG] and pregnancy-associated plasma protein-A (PAPP-A)) at 11–14 weeks of gestation. allowing the opportunity for further discussion before embarking on screening. 0. 251 women) and inhibin (three studies. There were no significant differences in weight-corrected free β-hCG (type 1 diabetes MoM 0.52).8% (23/42) of babies. should be given to women as early as possible and ideally before the booking visit. more than twice the expected rate. First-trimester screening for chromosomal anomalies The NICE antenatal care guideline9 recommends that all pregnant women should be offered screening for Down’s syndrome.03 for inhibin (MoM values are ratios of median MoM in women with type 1 diabetes to median MoM in women without diabetes). enabling families time to prepare.6% (48/67) of babies. The second-trimester ultrasound scan (at 18–20 weeks of) should not be routinely used for Down’s syndrome screening using soft markers. If a woman receives a screen-positive result.4-fold). one study has been published on free β-hCG.Diabetes in pregnancy was 46 per 1000 births in women with diabetes (48 per 1000 births for type 1 diabetes. Therefore.92 for AFP. [EL = 2+] Since publication of the meta-analysis.96 for free β-hCG and 1. namely the ‘triple test’ or ‘quadruple test’ (hCG. The screening test offered should be the ‘combined test’ (nuchal translucency (NT).2-fold) and congenital heart disease (3. 1350 women with type 1 diabetes). allowing antenatal treatment. she should have rapid access to appropriate counselling by trained staff. The integrated test should not be routinely used as a screening test for Down’s syndrome. uE3 (six studies. The presence of an increased nuchal fold or two or more soft markers should prompt the offer of fetal medicine referral.75 to 1.9 all pregnant women should be offered screening for congenital malformations at 18–20 weeks of gestation as part of routine antenatal care. The presence of an isolated soft marker with an exception of increased nuchal fold noted on the routine anomaly scan (at 18–20 weeks of gestation) should not be used to adjust the a priori risk for Down’s syndrome.87. [EL = 2+] Two studies were identified that compared levels of PAPP-A in women with and without diabetes during pregnancy. No CIs or tests of statistical significance were presented in the meta-analysis. One study compared 79 women with type 1 diabetes to 93 pregnant women without diabetes. [EL = 3–4] The benefits of screening for congenital malformations include the opportunity for counselling. Anomalies in 65% (71/109) of babies were diagnosed antenatally. 0. but provision should be made to allow later screening (up to 20 weeks of gestation) for women booking later in pregnancy. At 15–20 weeks of gestation the most clinically effective and cost-effective serum screening test should be offered. 445 women). [EL = 2+] 98 .

All six CNS abnormalities were detected antenatally. Seven defects occurred in women with type 1 diabetes. Other views did not contribute to detection of a defect.272 The ultrasound included four chambers of the heart and the great vessels. The majority (86% [19/22]) of repeat scans were also unsatisfactory.3%) were present in the fetuses of women with diabetes. 95% CI 0.275 There were 11 heart defects. 48 gestational diabetes) considered the utility of different echocardiogram views. PPV 92%.36). There were eight heart anomalies of which five were detected antenatally.5%. 47 type 2 diabetes.01. Of the 48 whose image quality was judged to be unsatisfactory there were eight major congenital anomalies. Views included the four chamber view. 52% (18/32) of which were detected antenatally. The two missed cases were in women who were obese. The incidence of major congenital malformations was greater in the women with diabetes than in the low-risk control group (8% versus 1.5%. The test performance of the standard four chamber view was: sensitivity 33%. PPV 100%. mainly because of maternal obesity (45/48). [EL = 2+] One study was identified that compared NT results in 195 women with type 1 diabetes to those in 33 301 women without diabetes. women without diabetes MoM 1. The detection rate was significantly lower in the women with diabetes (30% versus 73%. [EL = 2++] In a study of 289 women with diabetes273 comprehensive ultrasound including a four chamber view undertaken at 18 weeks of gestation by a perinatologist had a test performance for detection of non-cardiac anomalies as follows: sensitivity 59%. three in women with type 2 diabetes and one in a woman with insulin-requiring gestational diabetes. PPV 100%. A total of ten major anomalies (7. There were eight cardiac anomalies (3. nine of which were detected antenatally.05.269 There was no difference in mean NT between the two groups (0. Of the 82 women with diabetes who had satisfactory images. In comparison the test performance for echocardiogram was: sensitivity 92%.01) and the mean BMI was significantly higher (29 kg/m² versus 23 kg/ m²). specificity 99.001). 45 type 2 diabetes) with 12 169 low-risk pregnant women for the same period. two had congenital malformations. The test performance was: sensitivity 56%.02. [EL = 2++] A cohort study considered 432 women with type 1 diabetes who underwent ultrasound screening between 12 and 23 weeks of gestation. P < 0. NPV 99%.5%). The false positive was a case of perimembranous ventricular-septal defect.0002 mm.Antenatal care The second study compared PAPP-A levels in 79 women with type 1 diabetes to those in 16 366 women without diabetes. and longitudinal view of the aortic arch. specificity 100%. specificity 100%. the left ventricular long-axis view with visualisation of the aortic outflow tract. The two missed cases (pulmonary atresia with a 99 . six of which were detected antenatally by echocardiogram.271 All women had routine ultrasound at 16–24 weeks of gestation. At birth 32 babies had 38 major congenital malformations. specificity 99%. The lesions most commonly missed by sonography were ventricular septal defect. The test sensitivity was 85. Only one was detected antenatally (detection rate 12. [EL = 2++] A cohort study reported on 250 women274 with pre-existing diabetes who underwent fetal echocardiogram at 20–22 weeks of gestation. One fetus had an apparently normal heart at 21 weeks of gestation but was found to have a small atrial-septal defect at birth.7% and specificity was 99. The sensitivity of the four chamber view was 73% (8/11) and specificity was 100%.418). P = 0. P < 0.0358 mm versus 0.269 There was no significant difference in PAPP-A levels (type 1 diabetes MoM 1. There was one false-negative result and one false-positive result. NPV 97%.‘ A cohort study compared 130 women with diabetes (85 type 1 diabetes.4%. unilateral renal abnormality. Thirty-seven percent (48/130) of scans undertaken in women with diabetes were judged to be unsatisfactory. NPV 98%. [EL = 2++] Second-trimester ultrasound screening for structural anomalies The NICE antenatal care guideline9 recommends that ‘ultrasound screening for fetal anomalies should be routinely offered. and cleft palate without cleft lip. All examinations were undertaken by three experienced ultrasonographers. PPV 90%. NPV 97%.2%). [EL = 2++] A study of 223 women with insulin-requiring diabetes (128 type 1 diabetes.7%) and three minor anomalies (2. Both were detected antenatally (detection rate 100%). P = 0. the short-axis view with visualisation of the pulmonary outflow tract and ductus arteriosus. abnormal hand or foot.83 to 1. normally between 18 weeks 0 days and 20 weeks 6 days. The sensitivity of the four chamber view and aortic outflow tract was 82% (9/11) and the specificity was 100%.

the GDG considered that this was an area where a different screening programme from that used in routine antenatal care might be justified on health economic grounds. The indications for referral included pre-existing diabetes (without additional indication. No abnormal fetal echocardiograms were reported in women with isolated pre-existing diabetes (i. adjustments should continue to be applied. An economic model was used to compare the cost-effectiveness of screening for congenital cardiac malformations using the four chamber plus outflow tracts view versus the four chamber view alone. aneuploidy (14%) other fetal anomaly (17%) and fetal arrhythmia (3%). n = 226).e. fetal anomaly seen on anatomy ultrasound (n = 130) and family history of congenital heart disease (n = 133). [EL = 2++] Evidence statement A number of studies have found no significant differences between women with type 1 diabetes and women without diabetes in terms of NT and weight-corrected total hCG. The guideline recommends that all pregnant women should be offered screening for Down’s syndrome and that women should understand that it is their choice to embark on screening for 100 . Two studies have found conflicting results with regard to levels of PAPP-A. which represents current practice. No more anomalies are detected with additional views. The baseline model suggested that the four chamber plus outflow tracts view was highly cost-effective in pregnant women with diabetes with a cost per QALY of approximately £4. The indications for referral in these cases were an abnormal four chamber/left ventricular outflow tract view at ultrasound (66%).000. with a normal four chamber/left ventricular outflow tract view at ultrasound). Existing guidance As noted above. the NICE antenatal care guideline9 recommends that all pregnant women should be offered ultrasound screening for congenital malformations (ideally at 18–20 weeks of gestation) using the four chamber plus outflow tracts view as part of routine antenatal care. Therefore. [EL = 2++] One study276 examined 725 women (with or without diabetes) who had been referred for echocardiogram following a comprehensive anatomy ultrasound that included a four chamber/ left ventricular outflow tract view. A number of well-designed observational studies have found more congenital anomalies are detected antenatally in women with diabetes when antenatal examination includes views of the four chambers of the fetal heart and outflow tracts. A meta-analysis found weight-corrected AFP to be approximately 8% lower in women with type 1 diabetes and weight-corrected uE3 to be 6% lower in women with type 1 diabetes and therefore adjustments should be applied accordingly. One study found detection rates were significantly worse in women with diabetes compared with low-risk women. Twenty-nine echocardiograms were reported as abnormal (4%). Women should be given information regarding the purpose and implications of the anomaly scan in order to enable them make an informed choice as to whether or not to have the scan. β-hCG and inhibin. until further evidence is available. further details are provided in Appendix E. Women with diabetes are at increased risk of having a baby with a cardiac malformation (the risk being approximately five times that of the general maternity population).Diabetes in pregnancy ventricular septal defect and an isolated ventricular septal defect) could theoretically have been detected on the four chamber view. On this basis it can be advised that no adjustment is required in these biochemical markers when calculating risks for congenital abnormalities in the fetuses of women with diabetes. The methods and results from the health economic modelling are summarised here. Therefore. One-way sensitivity analysis showed that four chamber plus outflow tracts view continued to be cost-effective when parameter values were varied within plausible ranges. This was largely attributed to obesity in women with diabetes resulting in unsatisfactory images. Cost-effectiveness The effectiveness of methods of screening for congenital cardiac malformations in women with diabetes was identified by the GDG as a priority for health economic analysis.

The GDG’s view is that a specialist cardiac scan should be offered at 22 weeks of gestation only if the results of the four chamber plus outflow tracts view are abnormal or if there is a relevant history of cardiac malformations. How effective is transvaginal ultrasound for the detection of congenital malformations in women with diabetes and coexisting obesity? Why this is important Obstetric ultrasound signals are attenuated by the woman’s abdominal wall fat.Antenatal care Down’s syndrome. There may be additional benefits of screening not taken into account in the model. therefore. This is likely to bring a cost saving to the NHS because there is currently a tendency to offer a specialist cardiac scan to many women with diabetes. she should have rapid access to appropriate counselling by trained staff. However. These data. needed to evaluate the diagnostic accuracy and affect on pregnancy outcomes of vaginal ultrasound in women with diabetes and coexisting obesity. two clinical studies have reported conflicting results in terms of whether levels of PAPP-A in women with type 1 diabetes are lower than those in other women. For this reason the GDG identified screening for congenital cardiac malformations using the four chamber plus outflow tracts view as a key priority for implementation for women with diabetes (this form of screening is recommended as part of routine antenatal care but it does not form a key priority for implementation in the NICE antenatal care guideline9). Data from European Surveillance of Congenital Anomalies (EUROCAT) and published literature (see Appendix D) suggest that an antenatal diagnosis of TGA may reduce neonatal mortality. From evidence to recommendations A health economic model demonstrated the cost-effectiveness of screening for congenital cardiac malformations based on the four chamber view of the fetal heart and outflow tracts relative to current practice of screening using the four chamber view alone. the existence of which would tend to further improve the relative cost-effectiveness of screening based on the four chamber plus outflow tracts view. Further research is. together with the higher prevalence of cardiac malformations in pregnant women with diabetes compared to pregnant women without diabetes underpin this result. 101 . needed to evaluate the diagnostic accuracy and effect on pregnancy outcomes of screening tests for Down’s syndrome incorporating measurements of PAPP-A in women with pre-existing diabetes. therefore. If a woman receives a screen positive result. Research studies are. This is important because women with diabetes are at increased risk of having a baby with a congenital malformation. Research recommendations for screening for congenital malformations How reliable is first-trimester screening for Down’s syndrome incorporating levels of pregnancyassociated plasma protein (PAPP-A) in women with pre-existing diabetes? Why this is important Several screening tests for Down’s syndrome incorporate measurements of PAPP-A. and this may limit the sensitivity of abdominal ultrasound screening for congenital malformations. but there is currently no evidence that it is more effective than abdominal ultrasound. Vaginal ultrasound does not have this difficulty. Many women with diabetes (and particularly women with type 2 diabetes) are obese. Current practice is to adjust PAPP-A measurements in women with diabetes on the assumption that their PAPP-A levels are indeed lower than those of other women. Recommendations for screening for congenital malformations Women with diabetes should be offered antenatal examination of the four chamber view of the fetal heart and outflow tracts at 18–20 weeks.

280.281 Accuracy of estimated fetal weight is worse in women with diabetes282 and for macrosomic babies. Women with diabetes are also at risk of having a baby that is SGA. Fetal size can also be measured by sonography.286 Both large and normal weight babies of women with diabetes tended to have their 102 . Mean errors in estimating fetal weight are between 8% and 15% of actual birthweight.30 to 0. P = 0. and absolute error.285 One hundred and sixty-five women with pre-existing diabetes or gestational diabetes who had sonograms to estimate fetal weight after 36 weeks of gestation and within 2 weeks of birth were included in the study. A positive test result is more accurate for ruling in macrosomia than is a negative test for ruling it out.45). A systematic review of 63 studies (51 evaluating the accuracy of estimated fetal weight and 12 the accuracy of fetal abdominal circumference) involving 19 117 women pooled data to produce summary receiver operating characteristic (sROC) curves for studies with various test thresholds. The risks associated with a baby that is SGA are not as well documented as for macrosomia.9 (95% CI 5. The sROC curve area for estimated fetal weight was not different from the area for fetal abdominal circumference (0. biparietal diameter. Abdominal circumference also has the advantage of being a single measure that is accessible even when the head is engaged in the pelvis.1 and Chapter 7). There was no difference in accuracy between estimated fetal weight and abdominal circumference in the prediction of macrosomia at birth. the Hadlock formula279 uses femur length.91). for example. head circumference and abdominal circumference.7 (95% CI 4.48 (95% CI 0. [EL = 1++] A diagnostic accuracy study compared 31 published formulas for estimated fetal weight in predicting macrosomia (birthweight 4000 g or more) in babies of women with diabetes.6) for a positive test and 0. All 31 formulas for estimating fetal weight had similarly poor accuracy for prediction of macrosomia. The LRs suggest that both tests are only moderately useful at best. Macrosomia is a risk factor for shoulder dystocia. also referred to as LGA. [EL = 2+] A cohort study evaluated the reliability of ultrasound estimation of fetal weight in 1117 women (48 with gestational diabetes) with a singleton pregnancy who had undergone ultrasound estimation of fetal weight less than 7 days before a term birth (at or later than 37 weeks of gestation). LGA babies have accelerated growth of insulin-sensitive tissue such as abdominal wall fat.283 Compared with babies of women with diabetes who are AGA.278 Clinical assessment of fetal size is by measurement of the symphysis–fundal height.Diabetes in pregnancy 5. For ultrasound fetal abdominal circumference of 36 cm the LR for a positive test for predicting birthweight over 4000 g was 6.1). but at least one study was identified that suggested that babies who were SGA (< 10th percentile for gestational age) have an increased risk of perinatal morbidity and mortality. The two main ultrasonic methods for predicting birthweight are estimated fetal weight (EFW) and abdominal circumference of the fetus. asphyxia or prolonged labour. operative delivery and postpartum haemorrhage (see Section 6.85.277 There is no clear consensus for monitoring fetal size in pregnant women with diabetes.7 Monitoring fetal growth and wellbeing Description of the evidence Fetal growth Women with gestational diabetes and pre-existing diabetes are at increased risk of having a baby with macrosomia (see Sections 4. EFW uses a combination of parameters. brachial plexus injury.60) for a negative test.278 Summary likelihood ratios (LRs) for positive and negative test results were generated for an estimated fetal weight of 4000 g and an abdominal circumference of 36 cm for predicting birthweight over 4000 g.38 to 0.2 to 9. Macrosomia is defined in terms of absolute birthweight (usually more than 4000 g) or birthweight percentile for gestational age (usually ≥ 90th percentile). systematic error.3 to 7. Three measures of accuracy were compared: area under the ROC curve relating estimated fetal weight to macrosomia.37 (95% CI 0.279 EFW increases the rate of caesarean section in false positives (AGA babies incorrectly diagnosed as LGA).1 and 5.0) and the LR for a negative test was 0. For predicting a birthweight of over 4000 g the summary LR was 5.284 Abdominal circumference is therefore considered to be a more relevant measure of diabetes-related macrosomia and the risk of shoulder dystocia.87 versus 0.

74 to 0. Although women with diabetes are at increased risk for these conditions. fewer inductions of labour (OR 0. Adding ultrasound to clinical information improved prediction. Monitoring for fetal wellbeing assumes that fetal compromise can be identified and that appropriately timed intervention (induction of labour or caesarean section) may reduce the risk of perinatal morbidity.289 Clinical and ultrasound estimates were made at 28. hypocalcaemia. Prediction improved with closeness to birth.39. 48 gestational diabetes) compared the prediction power. Doppler ultrasound offers no benefits to the low-risk population. It is used during pregnancy to assess fetus–placenta and/or uterus–placenta circulation. There was no difference in the prediction power for macrosomia between clinical and ultrasound measurements.7 to 0. the majority of adverse outcomes in pregnancies complicated by diabetes are not associated with placental insufficiency. asphyxia and fetal death. An adverse pregnancy outcome was defined as a pregnancy in which the baby was born before 37 weeks of gestation or had at least one of the following: growth restriction. 95% CI −01. [EL = 2+] A prospective study of 181 women with diabetes (133 pre-existing type 1 diabetes. Given that reliability of ultrasound estimation of fetal weight to detect larger babies was poor. [EL = 2+] A retrospective cohort study involving 242 pregnant women with IGT evaluated the performance of estimated fetal weight and fetal growth velocity in the prediction of birthweight.292 [EL = 1++] Abnormal umbilical Doppler ultrasound results are associated with chronic placental insufficiency. Amniotic fluid insulin levels demonstrated a similar significant difference between women with pre-existing diabetes and those with gestational diabetes or IGT.83.56.298 In this study involving 207 women with diabetes.290 Doppler ultrasound: Doppler ultrasound uses sound waves to detect the movement of blood in the umbilical artery. A systematic review considered the effectiveness of Doppler ultrasound in high-risk pregnancies. These are umbilical artery Doppler ultrasound velocimetry. Estimated fetal weight and fetal growth velocity did not predict neonatal hypoglycaemia. 95% CI 0.43 to 2. [EL = 2++] Fetal wellbeing Three main tests are used by obstetricians to monitor fetal wellbeing.43 to 0. 103 . Doppler ultrasound in high-risk pregnancies (especially those complicated by hypertension or presumed impaired fetal growth) was associated with fewer perinatal deaths. The study demonstrated that ultrasound measures of fetal size and growth are not sufficiently accurate to predict those babies likely to be at risk from the effects of fetal hyperinsulinaemia. significantly greater mean estimated fetal weight and greater mean birthweight centile than those with gestational diabetes or IGT. at different gestational ages. but only to a small extent. [EL = 1++] However. 34 and 38 weeks of gestation or before birth.287 Babies of women with pre-existing diabetes had significantly greater mean growth velocity (1.293 Small studies of women with gestational diabetes or pre-existing diabetes with good glycaemic control have considered the performance of Doppler ultrasound in predicting any adverse pregnancy outcome and have reported low sensitivities294–297 [EL = 2++ to 2+] Nonetheless a study has reported that Doppler ultrasound is better than fetal cardiotocography or biophysical profile in pregnant women with diabetes. all three tests were performed concurrently within 1 week of birth. 95% CI 0. fetal cardiotocography (non-stress test) and the biophysical profile. of clinical and ultrasound measurements for fetal size.04).72) without adverse effects.288 The study showed that estimated fetal weight and fetal growth velocity have limited utility in predicting LGA babies. the study suggests that ultrasound use in the management of suspected macrosomia should be discouraged. admission to neonatal intensive care. The study found all measurements were poor predictors of eventual standardised birthweight. 95% CI 0.Antenatal care weight underestimated. Compared with no Doppler ultrasound.93) and fewer admissions to hospital (OR 0.23 versus 0. P = 0.39.95.291 The review included 11 studies involving 7000 women. as occurs in pregnancies complicated by pre-eclampsia and IUGR. [EL = 2+] A retrospective cohort study investigated the association between ultrasound fetal biometry and amniotic fluid insulin levels at birth in 93 pregnant women with pre-existing diabetes or IGT.

9 to 3. which may reflect a relative reduction in basal uteroplacental blood flow and the need for cautious interpretation of Doppler indices in these women.0 1. The babies demonstrated no evidence of fetal distress at birth and there was no relationship between the 104 .4 to 3. P < 0. IUGR and neonatal metabolic complications were also significantly correlated with elevated Doppler indices.5 65.3% 96.5 Performance of umbilical artery Doppler ultrasound. the clinical and Doppler parameters were significantly different in women with pre-eclampsia than in those without pre-eclampsia. The data indicate an increased resistance circuit among women with diabetes and vasculopathy.Diabetes in pregnancy hypoglycaemia. There was a weak positive linear correlation (r = 0.0% 65.9 to 2. [EL = 2+] A prospective cohort study investigated Doppler ultrasound measurement of the fetal umbilical artery velocimetry in 56 women with diabetes.2% 79.7 1. [EL = 1+] Sixty-five pregnant women with diabetes were examined in a cohort study to evaluate the clinical usefulness of Doppler ultrasound flow velocity waveform analysis in such pregnancies.3% 61.6 1. regardless of glycaemic control.49. specificity 92%. The study suggests that Doppler ultrasound flow velocity waveform analysis may be clinically useful only in pregnancies complicated by diabetes and coexisting pre-eclampsia. hyperbilirubinaemia.2% 69. respiratory distress syndrome or fetal risk requiring caesarean section. There was no difference in various clinical and Doppler ultrasound parameters between women with good glycaemic control and those with poor control.001 A prospective double-blind randomised study was performed between 28 and 40 weeks of gestation in 92 pregnant women with diabetes to evaluate a random single Doppler ultrasound measurement of the systolic : diastolic ratio of the umbilical artery as a predictor of perinatal outcome in pregnancies complicated by diabetes.5 < 0.7% 2. The performance of the three tests is summarised in Table 5.9 0. Statistically significant correlations were found between elevated Doppler indices and maternal vasculopathy associated with hypertension and worsening renal insufficiency.5 0.109 Umbilical artery Doppler ultrasound systolic : diastolic ratio ≥ 3 25. of whom 14 had varying degrees of vascular complications.4% 2.001).300 Umbilical and uterine artery flow velocity waveforms were obtained during the third trimester with a continuous wave Doppler ultrasound device.001 systolic : diastolic ratio ≥ 2.4% 49. There were no perinatal deaths in this series.02) between maternal HbA1c and umbilical artery flow velocity waveforms (systolic : diastolic ratio).30.5.6% 1. A tendency towards adverse outcomes was observed at systolic : diastolic ratios approaching 4.6% 55. NPV 86%.0% 97.0% 75. [EL = 2+] Another prospective cohort study was conducted to determine whether fetal aortic velocity waveforms were correlated with fetal outcome in pregnancies complicated by type 1 diabetes. The data suggest that the systolic : diastolic ratio of the umbilical artery offers no advantage over other well-established tests in the management of pregnancy in women with diabetes. P < 0. [EL = 2++] Table 5. The third-trimester systolic : diastolic ratio was greater than 3 in almost 50% of women with vasculopathy. fetal cardiotocography and biophysical profile in predicting overall adverse pregnancy outcome.0% 60. data from Bracero et al.3% 88.2 to 2. (1996)298 Cardiotocography non-reactive Sensitivity Specificity PPV NPV RR 95% CI P value 25.301 The mean Doppler ultrasound values were higher in women with diabetes and vasculopathy than in women without diabetes and women with diabetes but no vasculopathy. In contrast. Proteinuria correlated better with umbilical artery systolic : diastolic ratio (r = 0.0 < 0.9% 67.299 The performance of the Doppler ultrasound measurement as a predictor of poor perinatal outcome was: sensitivity 39%.009 Biophysical profile ≤ 6 8. PPV 54%.0% 1.7 0.302 Fetal aortic blood flow was assessed in 30 pregnant women with type 1 diabetes.

preterm deliveries. no clinical studies were identified that compared clinical outcomes using the two methods.31). 95% CI 0.4 to 6.9% of babies weighing less than 2500 g.9% of singleton babies born to women with type 1 diabetes and 1. there were also concerns about poor interpretation of ultrasound scans and about actions taken in response to tests.5 to 1. adjusted for maternal age and deprivation). 22. The GDG’s view was that three 105 . [EL = 3–4] The CEMACH enquiry found no difference in the proportion of women with type 1 or type 2 diabetes with antenatal evidence of macrosomia (P = 0. this is higher than the national average for England and Wales (0.8.3% singleton babies born to women with type 2 diabetes were less than 1000 g. Apgar scores at 1 minute and 5 minutes.5 per 100 births. [EL = 2+] Current practice The CEMACH enquiry found that 21% of singleton births with a known birthweight had a birthweight of 4000 g or more in women with poor pregnancy outcomes. rather than diabetes per se.7% of babies 2500–3999 g.33 This was higher than the national average of 11%. The CEMACH enquiry found that Erb palsy occurred in 4.3. For macrosomic babies.4 to 12.33 Fetal surveillance was sub-optimal for 20% of 37 babies with antenatal evidence of fetal growth restriction and for 45% of 129 babies with antenatal evidence of macrosomia. but its effectiveness is limited to high-risk pregnancies defined in terms of IUGR and/or pre-eclampsia. fetal abdominal circumference alone versus abdominal circumference plus fetal head circumference) precluded formal cost-effectiveness analysis. However.3. 95% CI 2. Umbilical artery Doppler ultrasound has better diagnostic accuracy as a test of fetal wellbeing in pregnant women with diabetes than has fetal cardiotocography or biophysical profile. birthweight.2 [EL = 3] The CEMACH enquiry found 0. A total of 5. Additional case–control analysis showed an association with fetal and neonatal death after 20 weeks of gestation. The GDG’s discussions included consideration of the frequency of ultrasound assessment of fetal growth and the implications for cost-effectiveness.3.2 [EL = 3] The main concerns of the enquiry panels regarding surveillance of macrosomic and growthrestricted babies was lack of timely follow-up (affecting approximately 80% of babies).0% of babies 4000–4249 g. For babies with antenatal evidence of macrosomia. [EL = 3–4] Evidence statement The main ultrasonic methods for predicting birthweight (EFW and abdominal circumference) perform similarly in terms of diagnostic accuracy in women with diabetes. but not with fetal anomaly. but antenatal evidence of fetal macrosomia was not (OR 0.9% of babies 4500 g or more. The rate of shoulder dystocia was related to birthweight with 0. 25% of babies 4250–4499 g and 42. Doppler ultrasound is also a better predictor of adverse maternal and neonatal outcomes.0.9.5%). adjusted for maternal age and deprivation). sub-optimal fetal surveillance was associated with poor pregnancy outcome (OR 5.99) or fetal growth restriction (P = 0.7% births were severely macrosomic singleton births (a birthweight of 4500 g or over). adjusted for maternal age and deprivation). this is greater than the incidence of 0. The CEMACH enquiry case–control study reported that antenatal evidence of fetal growth restriction was associated with poor pregnancy outcome (OR 2.9% of vaginal births. The lack of comparative data in relation to clinical outcomes resulting from ultrasonic methods for assessing fetal growth (for example.Antenatal care mean third-trimester fetal aortic systolic : diastolic ratios and perinatal death. Cost-effectiveness The effectiveness of methods for monitoring fetal growth and wellbeing in women with diabetes was identified by the GDG as a priority for health economic analysis. or neonatal metabolic abnormalities. The data demonstrate a poor correlation between fetal aortic Doppler waveform analysis and fetal outcome. 95% CI 1.42 per 1000 live births reported in the general population. 4.33 [EL = 3–4] The CEMACH enquiry found that shoulder dystocia was documented in 7.

In making this recommendation the GDG sought to effect a change in clinical practice that would bring a cost saving to the NHS. Ultrasound estimation of fetal size for suspected LGA unborn babies should not be undertaken in a low-risk population. The clinical evidence in relation to monitoring fetal wellbeing showed that umbilical artery Doppler ultrasound is more effective in predicting adverse outcomes in women with diabetes than fetal cardiotocography or biophysical profile. even in women with gestational diabetes controlled by diet alone) because of the increased risk of macrosomia. From evidence to recommendations In the absence of comparative data on the effectiveness of different methods of ultrasound monitoring of fetal growth. Routine monitoring of fetal wellbeing before 38 weeks is not recommended in pregnant women with diabetes. it would also allow assessment of the need for. and that current practice involves routine use of Doppler ultrasound to monitor fetal wellbeing in women with diabetes. rather than providing estimates of fetal growth that might be masked by measurement error. Given that the clinical effectiveness of Doppler ultrasound is limited to women with other risk factors (notably IUGR and/or pre-eclampsia). and response to. the GDG recommended that fetal growth and amniotic fluid volume (to detect polyhydramnios) should be monitored by ultrasound every 4 weeks from 28 weeks of gestation to 36 weeks of gestation. bring a cost saving to the NHS. Existing guidance The NICE antenatal care guideline9 recommends that symphysis–fundal height should be measured and recorded for pregnant women at each antenatal appointment from 24 weeks of gestation. Evidence shows that monitoring for fetal wellbeing using umbilical artery Doppler ultrasound is a better predictor of pregnancy outcome than fetal cardiotocography and biophysical profile in women with diabetes. bring a cost saving to the NHS. 106 . insulin therapy. routine monitoring of fetal wellbeing for women with diabetes is not recommended before 38 weeks of gestation because the effectiveness of Doppler ultrasound is limited to women at risk of IUGR and/or pre-eclampsia. therefore. a recommendation not to monitor fetal wellbeing routinely before 38 weeks of gestation was considered likely to be cost-effective. A fetal growth scan to detect SGA unborn babies should be offered to women if the symphysis–fundal height measurement is at least 3 cm less than the gestational age in weeks. Doppler ultrasound should not be used to monitor fetal growth during pregnancy. Nevertheless. three scans at 4-weekly intervals starting at 28 weeks of gestation was thought to represent a reduction in the frequency of growth scans compared with current clinical practice that would. unless there is a risk of intrauterine growth restriction.e. Customised fetal growth charts should not be used for screening for SGA babies. Recommendations for monitoring fetal growth and wellbeing Pregnant women with diabetes should be offered ultrasound monitoring of fetal growth and amniotic fluid volume every 4 weeks from 28 to 36 weeks. However. Women with diabetes and a risk of intrauterine growth restriction (macrovascular disease and/or nephropathy) will require an individualised approach to monitoring fetal growth and wellbeing. Fetal growth and amniotic fluid volume should be measured in all women with pre-existing diabetes and gestational diabetes (i. therefore. The GDG’s view is that this would represent a change in clinical practice which would effect a reduction in the frequency of monitoring for fetal growth and amniotic fluid volume in women with diabetes and would.Diabetes in pregnancy scans should be offered (rather than two): this would allow healthcare professionals to advise women with diabetes on the direction of pregnancy.

Wales and Northern Ireland in relation to antenatal care. a midwife and a dietitian.33 [EL = 3–4] The CEMACH enquiry panels commented that infrequent clinic appointments.7) and management of maternal risks.8 Timetable of antenatal appointments Description of the evidence No specific searches were undertaken for this section of the guideline. but there is currently insufficient clinical evidence to evaluate them. a diabetes specialist nurse. a diabetes specialist midwife and a dietitian. The two most frequently cited categories for sub-optimal antenatal care were fetal surveillance (monitoring fetal growth and wellbeing. adjusted for maternal age and deprivation). a diabetes physician. There had been an increase in provision of staff over the preceding 8 years. a diabetes specialist nurse. Well-designed randomised controlled trials that are sufficiently powered are needed to determine whether these approaches are clinically and cost-effective. as a minimum:33 • targets for glycaemic control • a schedule for retinal screening • a schedule for renal screening • a plan for fetal surveillance during birth • postnatal diabetes care. a diabetes physician.2 to 2. 5. including recommendations about what should happen at each appointment. Current practice The CEMACH diabetes in pregnancy programme provides data on current practice in England. for women with type 1 and type 2 diabetes. Seventy-five percent of women were reported to have maternity and diabetes care provided in a joint clinic. although only 22% of women were reported to have the entire multidisciplinary team involved in their care. [EL = 3–4] Existing guidance The NSF for diabetes20 recommends that antenatal care for women with diabetes should be delivered by a multidisciplinary team consisting of an obstetrician.33 [EL = 3–4] The CEMACH enquiry recommended that an individualised care plan for pregnancy (and the postnatal period) be used and that the care plan should include. cardiotocography and other biophysical tests) have been shown to have poor sensitivity for predicting such events. see Section 5. 107 . including care plans. Conventional tests of fetal wellbeing (umbilical artery Doppler ultrasound. lack of multidisciplinary involvement and communication issues were factors in sub-optimal diabetes care in pregnancy in some of the women in the case–control study. It was recommended that the care plan should be implemented from the beginning of pregnancy by a multidisciplinary team present at the same time in the same clinic. The enquiry panels classified maternity care during pregnancy as sub-optimal for 58% of women who had poor pregnancy outcomes and 44% of women who had good pregnancy outcomes (OR 1. Other categories cited included problems with the antenatal diabetes multidisciplinary team. The NICE antenatal care guideline9 contains recommendations on the schedule of appointments that should be offered as part of routine antenatal care. with the availability of a diabetes specialist midwife in the antenatal clinic increasing from 25% to 77% of units. The evidence is drawn from publications identified in searches for other sections.33 [EL = 3–4] The enquiry reported that 63% of maternity units in England Wales and Northern Ireland had a full multidisciplinary team comprising an obstetrician. and the availability of a dietitian increasing from 40% to 80% of units. 95% CI 1.8.Antenatal care Research recommendations for monitoring fetal growth and wellbeing How can the fetus at risk of intrauterine death be identified in women with diabetes? Why this is important Unheralded intrauterine death remains a significant contributor to perinatal mortality in pregnancies complicated by diabetes.9. There were no significant differences between women with type 1 and type 2 diabetes in terms of sub-optimal antenatal care. Alternative approaches that include measurements of liquor erythropoietin and magnetic resonance imaging spectroscopy may be effective.

4). Women with pre-existing diabetes may already have attended for preconception care and advice.3 and 5. but methods for surveillance (testing for proteinuria) and management of pre-eclampsia in women with diabetes are no different to those for women without diabetes. Women with pre-existing diabetes who have not had a retinal assessment in the previous 12 months should be offered an assessment at the first presentation in pregnancy (see Section 5.6. medications for diabetes and its complications should also be reviewed at this time.1).1) and because diabetes can disrupt the menstrual cycle leading to difficulty in determining the timing of ovulation. The GDG’s discussions included consideration of screening for Down’s syndrome. the first antenatal appointment provides an opportunity to reinforce information. The 108 .9 except where specific additions and/or differences are indicated below to support the recommendations made elsewhere in the guideline.4 and 5. First antenatal appointment The GDG’s view is that women with diabetes should be offered confirmation of viability and gestational age at the first antenatal appointment. The GDG’s discussions also included consideration of surveillance for pre-eclampsia. Ongoing opportunites for accessing information. 20 weeks of gestation Women with diabetes should be offered an ultrasound anatomical examination of the four chamber view of the fetal heart and outflow tracts at 20 weeks of gestation because the diagnostic accuracy is better at 20 weeks of gestation than at 18–19 weeks of gestation (see Section 5.6).5). Evidence from the CEMACH enquiry shows that many maternity units have not yet implemented the recommendation in the NSF for diabetes to provide diabetes and maternity care in a joint diabetes/antenatal clinic delivered by a multidisciplinary team.Diabetes in pregnancy From evidence to recommendations The GDG’s view is that women with diabetes who are pregnant should be offered immediate contact with a joint diabetes and antenatal clinic. The schedule of appointments for routine antenatal care recommended in the NICE antenatal care guideline9 includes testing urine for proteinuria at every appointment. All women with diabetes should have contact with the diabetes care team for assessment of glycaemic control every 1–2 weeks throughout the antenatal period (this could include telephone contact) and HbA1c should be used to assess long-term glycaemic control in the first trimester of pregnancy (see Section 5. education and advice in relation to achieving optimal glycaemic control (including dietary advice). Women with diabetes are at increased risk of pre-eclampsia (see Sections 4.5). and so the GDG made no specific recommendations in relation to surveillance for pre-eclampsia. and so the GDG made no specific recommendations in relation to the schedule for screening for Down’s syndrome. Screening methods for Down’s syndrome in women with diabetes are currently no different to those for women without diabetes. The timing and content of antenatal care appointments for women with diabetes should follow the schedule for routine antenatal care appointments recommended in the NICE antenatal care guideline. and they should have contact with the diabetes care team for assessment of glycaemic control every 1–2 weeks throughout pregnancy. 16 weeks of gestation If any retinopathy is present at booking an additional assessment should be made at 16–20 weeks of gestation for women with pre-existing diabetes (see Section 5. education and advice should be offered to women with diabetes throughout the antenatal period. The main differences between routine antenatal care (as specified in the NICE antenatal care guideline)9 and antenatal care for women with diabetes are summarised in Table 5. Women who have not attended for preconception care and advice should be offered the corresponding information. education and advice for the first time. Women with pre-existing diabetes who have not had a renal assessment in the previous 12 months should be offered an assessment at the first presentation in pregnancy (see Section 5. a clinical history should seek to establish the extent of diabetes-related complications (including neuropathy and vascular disease).4). For these women. In formulating its recommendations the GDG sought to reinforce the recommendation contained in the NSF for diabetes. This is earlier than in routine antenatal care because diabetes is associated with a high rate of miscarriage (see Sections 3.

1).2).3 and the NICE antenatal care guideline9). the routine investigations that would normally be offered to nulliparous pregnant women at 31 weeks of gestation should instead be offered at 32 weeks of gestation because it is more convenient for the woman to have all the investigations at one visit. initiation of breastfeeding and the effect of breastfeeding on glycaemic control (see Section 8. Women who have been diagnosed with gestational diabetes as a result of routine antenatal screening enter the care pathway at 28 weeks of gestation (see Section 4. mode and management of labour and birth.Antenatal care GDG’s view is that the routine ultrasound scan for detecting structural anomalies.7). detection and management of neonatal hypoglycaemia and other diabetes-related complications (see Chapter 7). It is the GDG’s view that.2). However. 34 weeks of gestation No evidence was identified to suggest that antenatal care for women with diabetes should be different to routine antenatal care at 34 weeks of gestation. 32 weeks of gestation Ultrasound monitoring of fetal growth and amniotic fluid volume should be offered at 32 weeks of gestation as part of 4-weekly monitoring (see Section 5. 25 weeks of gestation No evidence was identified to suggest that antenatal care for women with diabetes should be different to routine antenatal care at 25 weeks of gestation. 38 weeks of gestation Induction of labour. 36 weeks of gestation Ultrasound monitoring of fetal growth and amniotic fluid volume should be offered at 36 weeks of gestation as part of 4-weekly monitoring (see Section 5. There was no reason to suppose that women with diabetes required testing for thyroid function.1). should be offered to women with diabetes at 38 weeks of gestation (see Section 6. which should be offered to all pregnant women. They should be offered information about the risks to the woman and the baby that is offered to women with pre-existing diabetes in the preconception period. Monitoring of fetal wellbeing should be offered to women with diabetes who are awaiting spontaneous labour.1). evidence shows that many women with diabetes give birth before 40 weeks of gestation (see Section 6.1). Women with pre-existing diabetes who had no diabetic retinopathy at their first antenatal clinic visit should be offered retinal assessment at 28 weeks of gestation (see Section 5. 109 .7). for women with diabetes. 39–41 weeks of gestation No evidence was identified to suggest that antenatal care for women with diabetes who have not given birth by 40 weeks of gestation should be different to routine antenatal care at 40– 41 weeks of gestation. elective induction of labour or elective caesarean section to reduce the risk of stillbirth and birth trauma associated with fetal macrosomia (see Section 6. 28 weeks of gestation Ultrasound monitoring of fetal growth (to detect LGA or SGA babies) and amniotic fluid volume (to detect polyhydramnios) should start at 28 weeks of gestation and continue at 4-weekly intervals (i. The information and advice should cover: timing. including early feeding. the GDG’s view is that women with diabetes should be offered information and advice in relation to intrapartum care and postnatal care at 36 weeks of gestation. management of the baby after birth.1).2). should also be performed at 20 weeks of gestation in women with diabetes because it is more convenient for the woman to have both scans at one visit. Evidence shows that women with diabetes are likely to give birth soon after 36 weeks of gestation. or caesarean section if indicated.4). see Section 5.e. either through spontaneous labour. changes to hypoglycaemic therapy during and after birth (see Sections 6.3 and 8. analgesia and anaesthesia (see Section 6.7). including options for elective early birth (see Section 6. and information about contraception and follow-up (see Section 8. Given the evidence. 32 weeks and 36 weeks. The GDG’s discussions also included consideration of thyroid function in women with diabetes. Monitoring of fetal wellbeing should be offered to women with diabetes who are awaiting spontaneous labour at 39–41 weeks of gestation.

rubella susceptibility and syphilis • offer screening for asymptomatic bacteriuria • offering screening for Down’s syndrome • offer early ultrasound scan for gestational age assessment • offer ultrasound screening for structural anomalies (18–20 weeks) • measure BMI. Early testing of glood glucose or OGTT for women with a history of gestational diabetes and/or ongoing IGT (18–20 weeks). hepatitis B virus. including discussion of the routine anomaly scan. offer verbal information supported by written information (on topics such as diet and lifestyle considerations. education and advice for the first time. anaemia. Additional/different care for women with diabetes First appointment (joint diabetes and antenatal clinic) • if the woman has been attending for preconception care and advice. For a woman whose placenta is found to extend across the internal cervical os at this time. with an opportunity to discuss issues and ask questions. offer verbal information supported by antenatal classes and written information. if the woman chooses. 28 weeks The next appointment for all pregnant women should occur at 28 weeks. for women who choose to have screening. 28 weeks Start ultrasound monitoring of fetal growth and amniotic fluid volume Retinal assessment for women with preexisting diabetes if no diabetic retinopathy was present at the first antenatal clinic visit. 16 weeks Retinal assessment for women with preexisting diabetes if diabetic retinopathy was present at booking (16–20 weeks). with an opportunity to discuss issues and ask questions. 18–20 weeks At 18–20 weeks. take clinical history to establish extent of diabetes-related complications (including neuropathy and vascular disease). 25 weeks No additional or different care for women with diabetes. another scan at 36 weeks should be offered and the results of this scan reviewed at the 36 week appointment. HIV. hepatitis B virus. HIV. blood pressure and test urine for proteinuria. 16 weeks The next appointment should be scheduled at 16 weeks to: • review. At the first (and possibly second) appointment. red-cell alloantibodies. the following tests should be arranged: • blood tests (for checking blood group and RhD status and screening for haemogloinopathies. pregnancy care services available. 25 weeks At 25 weeks of gestation.6 Timetable for antenatal appointments for women with diabetes Routine antenatal care (NICE antenatal care guideline)9 First appointment (booking) • give information. with an opportunity to discuss issues and ask questions. education and advice in relation to achieving optimal glycaemic control (including dietary advice) • if the woman has not attended for preconception care and advice give information. 110 . reassess planned pattern of care for the pregnancy and identify women who need additional care • investigate a haemoglobin level of less than 11 g/100 ml and consider iron supplementation if indicated • measure blood pressure and test urine for proteinuria • give information. At this appointment: • offer a second screening for anaemia and atypical red-cell alloantibodies • investigate a haemoglobin level of less than 10. discuss and record the results of all screening tests undertaken. continue to provide information. an ultrasound scan should be performed for the detection of structural anomalies. if indicated • offer anti-D to rhesus-negative women • measure blood pressure and test urine for proteinuria • measure and plot symphysis–fundal height • give information.Diabetes in pregnancy Table 5. Women diagnosed with gestational diabetes as a result of routine antenatal screening enter the care pathway. rubella susceptibility and syphilis) ideally before 10 weeks • urine tests (to check for proteinuria and screen for asymptomatic bacteriuria) • ultrasound scan to determine gestational age using: – crown–rump measurement if performed at 10 weeks 0 days to 13 weeks 6 days – head circumference if crown–rump length is above 84 millimetres • Down’s syndrome screening using: – nuchal translucency at 11 weeks 0 days to 13 weeks 6 days – serum screening at 15 weeks 0 days to 20 weeks 0 days. At this appointment: • measure and plot symphysis–fundal height • measure blood pressure and test urine for proteinuria • give information. another appointment should be scheduled for nulliparous women. anaemia. and review medications for diabetes and its complications • retinal assessment and renal assessment for women with pre-existing diabetes if these have not been undertaken in preceding 12 months • contact with the diabetes care team every 1–2 weeks throughout pregnancy for all women with diabetes and assessment of long-term glycaemic control using HbA1c (first trimester only).5 g/100 ml and consider iron supplementation. red-cell alloantibodies. with an opportunity to discuss issues and ask questions. 20 weeks Ultrasound scan for detecting structural anomalies and anatomical examination of the four chamber view of the fetal heart plus outflow tracts. offer verbal information supported by antenatal classes and written information • screening for gestational diabetes. offer verbal information supported by antenatal classes and written information. maternity benefits and sufficient information to enable informed decision making about screening tests) • identify women who may need additional care and plan pattern of care for the pregnancy • check blood group and RhD status • offer screening for haemoglobinopathies.

36 weeks At 36 weeks. 38 weeks Another appointment at 38 weeks will allow for: • measurement of blood pressure and urine testing for proteinuria • measurement and plotting of symphysis–fundal height • information giving. Monitoring of fetal wellbeing if baby not yet born. offer external cephalic version • review ultrasound scan report if placenta extended over the internal cervical os at previous scan • give information. discuss and record the results of screening tests undertaken at 28 weeks. breastfeeding. reassess planned pattern of care for the pregnancy and identify women who need additional care. IGT = impaired glucose tolerance. offer verbal information supported by antenatal classes and written information. 40 weeks No additional or different care for women with diabetes. Offer information and advice about timing. with an opportunity to discuss issues and ask questions. reassess planned pattern of care for the pregnancy and identify women who need additional care. All routine investigations normally scheduled for 31 weeks should also be conducted at 32 weeks. management of the baby after birth and information about follow-up. including options for management of prolonged pregnancy. discuss and record the results of screening tests undertaken at 28 weeks. offer verbal information supported by antenatal classes and written information. with an opportunity to discuss issues and ask questions. offer verbal information supported by antenatal classes and written information • review. 39 weeks Monitoring of fetal wellbeing if baby not yet born. including anaesthetic review/assessment. OGTT = oral glucose tolerance test. 34 weeks At 34 weeks. otherwise await spontaneous labour. with an opportunity to discuss issues and ask questions. with an opportunity to discuss issues and ask questions. Monitoring of fetal wellbeing if baby not yet born. or caesarean section if indicated. verbal information supported by written information. all pregnant women should be seen in order to: • offer a second dose of anti-D to rhesus-negative women • measure blood pressure and test urine for proteinuria • measure and plot symphysis–fundal height • give information. with an opportunity to discuss issues and ask questions. Plan postnatal diabetes management. mode and management of labour and birth. Monitoring of fetal wellbeing if baby not yet born. a The NICE clinical guideline on induction of labour is being updated and is expected to be published in June 2008. offer verbal information supported by antenatal classes and written information • review. 111 . Additional/different care for women with diabetes 32 weeks (not 31 weeks) Ultrasound monitoring of fetal growth and amniotic fluid volume. an appointment at 40 weeks should be scheduled to: • measure blood pressure and test urine for proteinuria • measure and plot symphysis–fundal height • give information. with an opportunity to discuss issues and ask questions. 41 weeks No additional or different care for women with diabetes. all pregnant women should be seen again to: • measure blood pressure and test urine for proteinuria • measure and plot symphysis–fundal height • check position of baby • for women whose babies are in the breech presentation. 38 weeks Induction of labour. verbal information supported by antenatal classes and written information. 34 weeks No additional or different care for women with diabetes 36 weeks Ultrasound monitoring of fetal growth and amniotic fluid volume. 41 weeks For women who have not given birth by 41 weeks: • a membrane sweep should be offereda • induction of labour should be offereda • blood pressure should be measured and urine tested for proteinuria • symphysis–fundal height should be measured and plotted • information should be given.Antenatal care Routine antenatal care (NICE antenatal care guideline)9 31 weeks Nulliparous women should have an appointment scheduled at 31 weeks to: • measure blood pressure and test urine for proteinuria • measure and plot symphysis–fundal height • give information. 40 weeks For nulliparous women.

Offer retinal assessment to women with pre-existing diabetes who showed no diabetic retinopathy at their first antenatal clinic visit. birth and early parenting (such as breastfeeding and initial care of the baby). Offer ultrasound monitoring of fetal growth and amniotic fluid volume. 112 . There were no research recommendations relating to the timetable of antenatal appointments for women with diabetes. 16 weeks 20 weeks 28 weeks Offer retinal assessment at 16–20 weeks to women with pre-existing diabetes who showed signs of diabetic retinopathy at the first antenatal appointment. available from www. Offer induction of labour. Offer ultrasound monitoring of fetal growth and amniotic fluid volume. Offer information and advice about: • timing. and start regular tests of fetal wellbeing for women with diabetes who are awaiting spontaneous labour. Offer tests of fetal wellbeing.7 describes where care for women with diabetes differs from routine antenatal care. Antenatal appointments for women with diabetes should provide care specifically for women with diabetes.Diabetes in pregnancy Recommendations for timetable of antenatal appointments Women with diabetes who are pregnant should be offered immediate contact with a joint diabetes and antenatal clinic. advice and support in relation to optimising glycaemic control. Offer four chamber view of the fetal heart and outflow tracts plus scans that would be offered at 18–20 weeks as part of routine antenatal care. but with the exception of the appointment for nulliparous women at 31 weeks. Offer tests of fetal wellbeing. Table 5. Review medications for diabetes and its complications. mode and management of birth • analgesia and anaesthesia • changes to hypoglycaemic therapy during and after birth • management of the baby after birth • initiation of breastfeeding and the effect of breastfeeding on glycaemic control • contraception and follow-up. Table 5. At each appointment women should be offered ongoing opportunities for information and education. Women with diabetes should have contact with the diabetes care team for assessment of glycaemic control every 1–2 weeks throughout pregnancy. including appointments at 25 weeks (for nulliparous women) and 34 weeks. Offer ultrasound monitoring of fetal growth and amniotic fluid volume. Take a clinical history to establish the extent of diabetes-related complications. education and advice about how diabetes will affect the (ideally by 10 weeks) pregnancy.7 Specific antenatal care for women with diabetes Care for women with diabetes during pregnancya Offer information.nice. Confirm viability of pregnancy and gestational age. Offer tests of fetal wellbeing.uk/CG062). in addition to the care provided routinely for healthy pregnant women (see ‘Antenatal care: routine care for the healthy pregnant woman’ [NICE clinical guideline 62].org. Offer to nulliparous women all investigations that would be offered at 31 weeks as part of routine antenatal care. 32 weeks 36 weeks 38 weeks 39 weeks 40 weeks 41 weeks a Women with diabetes should also receive routine care according to the schedule of appointments in ‘Antenatal care: routine care for the healthy pregnant woman’ (NICE clinical guideline 62). Offer retinal and/or renal assessment if these have not been undertaken in the previous 12 months. or caesarean section if indicated. Appointment First appointment (joint diabetes and antenatal clinic) 7–9 weeks Booking appointment Discuss information.

05 for days 2 and 3). 12. eight of whom were treated before the introduction of the algorithm (cohort 1) and another eight who were treated after its introduction (cohort 2). Women in cohort 1 had insulin doses adjusted individually based on the level of blood glucose obtained. The median blood glucose over the 5 days was 6. Regression analysis showed that HbA1c was the strongest predictor of preterm birth from 6–32 weeks of gestation and that the risk of preterm birth was more than 40% when HbA1c was above 7. They may help to delay birth and allow women to complete a course of antenatal steroids. all insulin doses should be increased by 20% • on day 5. [EL = 2+] The second study reported on the use of a supplementary intravenous sliding scale to indicate the required dosage of supplementary insulin infusion in six women receiving antenatal steroids. 8. [EL = 3] Tocolytic agents Tocolytic agents are used to inhibit uterine contractions.3 mmol/litre.305 The supplementary insulin was in addition to the woman’s usual subcutaneous insulin regimen and usual dietary programme.7 mmol/litre and 7.7 mmol/litre. The first study reported on a test of an algorithm for improved subcutaneous insulin treatment during steroid treatment (intramuscular administration of betamethasone 12 mg repeated after 24 hours). 9. Two studies were identified that reported on approaches to increasing insulin in women with diabetes undergoing treatment with antenatal steroids. 113 .7 mmol/litre. Data were collected on six women receiving dexamethasone. The overall rate of preterm birth was 23%.4 mmol/litre in cohort 2 (P < 0. If blood glucose levels were too high on the initial regimen (glucose 10. 14. 7. Seventy-five percent of all glucose measurements were within 4–10 mmol/litre.0 mmol/litre and 7.7 mmol/litre in cohort 1 and 7. women who experienced preterm birth had higher HbA1c throughout pregnancy. Betamimetics have been widely used for tocolysis. insulin dose and albumin excretion rate measured at 12 weeks of gestation and every second week thereafter were recruited and followed. The use of steroids in women with diabetes is associated with a significant worsening of glycaemic control requiring an increase in insulin dose. [EL = 2+] Antenatal steroids Antenatal steroids are administered to women who have a spontaneous or planned preterm birth to accelerate fetal lung development and prevent respiratory distress syndrome. the night insulin dose should be increased by 25% • on day 2.1 mmol/litre or more for 2 consecutive hours) the dosage regimen was moved up to the next level. The study involved 16 women. None of the women developed ketoacidosis or severe hypoglycaemia.9 Preterm labour in women with diabetes Description of the evidence Incidence of preterm birth A prospective cohort study303 examined the importance of glycaemic control and risk of preterm birth in women with type 1 diabetes who have normoalbuminuria and no pre-eclampsia during pregnancy. all insulin doses should be increased by 10–20% • during days 6 and 7. The additional infusion was commenced immediately before the first steroid injection and continued for at least 12 hours after the second injection. Seventy-one women with complete data on HbA1c.Antenatal care 5.7% at 8 weeks of gestation. all insulin doses should be increased by 40% • on day 3. If the blood glucose level was less than 4 mmol/litre the dosage regimen was reduced by one level.3 mmol/litre. all insulin doses should be increased by 40% • on day 4. range 32–88 U) in order to achieve glucose control following administration of dexamethasone.304 The algorithm was as follows: • on day 1 (the day on which the first betamethasone injection is given). Significant amounts of supplementary intravenous insulin were required (median dose 74 U. 7. the insulin doses should be gradually reduced to their levels before treatment.2 mmol/litre.6 mmol/litre.

310–314 Current practice CEMACH undertook a descriptive study of all pregnancies of women with pre-existing diabetes who gave birth or booked between 1 March 2002 and 28 February 2003. The most common reason given for non-administration of antenatal steroids was birth of the baby before the full course could be given. Of the remaining 293 women. 70. Recommendations for preterm labour in women with diabetes Diabetes should not be considered a contraindication to antenatal steroids for fetal lung maturation or to tocolysis. Women with insulin-treated diabetes who are receiving steroids for fetal lung maturation should have additional insulin according to an agreed protocol and should be closely monitored. Betamimetic drugs should not be used for tocolysis in women with diabetes.3).2 Of the 3474 women in this study with a continuing pregnancy at 24 weeks of 328 gave birth before 34 weeks of gestation. Evidence shows that administration of betamimetics to suppress labour induces hyperglycaemia and ketoacidosis. From evidence to recommendations The evidence supports the use of increased insulin dose immediately before and during antenatal administration of steroids.Diabetes in pregnancy although they are no longer recommended as the first choice for general use. 114 . There were no research recommendations relating to preterm labour in women with diabetes. However. Since the two protocols that have been evaluated were only moderately successful in achieving glycaemic control. Thirty-five of these pregnancies resulted in a stillbirth. [EL = 3] Evidence statement The use of antenatal steroids for fetal lung maturation in women with diabetes is associated with a significant worsening of glycaemic control. In a small group of women diabetes was considered a contraindication to antenatal steroid use. the two protocols evaluated were only moderately successful in keeping blood glucose levels at the desired level (less than 7 mmol/litre). women receiving additional insulin during administration of antenatal steroids should be closely monitored according to an agreed protocol in case the insulin dose requires further adjustment.306 Betamimetics increase blood glucose concentrations307–309 and several cases of ketoacidosis have been reported in women with diabetes following administration of these drugs (see Section 5. When tocolysis is indicated in women with diabetes an alternative to betamimetics should be used to avoid hyperglycaemia and ketoacidosis. Two studies that reported on approaches to modifying insulin dose in women undergoing antenatal steroid treatment showed that glycaemic control could be improved by increasing the insulin dose immediately prior to and during administration of antenatal steroids.3% received a full course of antenatal steroid therapy.

01).0001).0001).7% in the women without diabetes. Treatment of gestational diabetes reduced rates of macrosomia (more than 4000 g) to 10.317 The study found that 110 women had vaginal births and 56 had caesarean section. fetal position and fetal fat were factors associated with caesarean section.6% compared with 29. hypertension (P < 0. false-positive gestational diabetes (n = 580).0001) and shoulder dystocia (P < 0. pre-eclampsia (P < 0.8).5% compared with 28. However. P = 0.319 The rate of injury to the baby during birth was 1. insulin-controlled gestational diabetes (P < 0.2%. However.6% in the untreated group and 20. [EL = 3] A prospective case–control study (n = 3778) from Canada examined the relationship between caesarean section rates and gestational glucose intolerance (3 hour 100 g OGGT). 95% CI 2.1 Intrapartum care Timing and mode of birth Description of the evidence Eight epidemiological studies were identified that examined the effect of diabetes on timing and mode of birth.0001). Women with gestational diabetes had higher rates of macrosomia (28. Three studies were identified that examined vaginal birth after previous caesarean section (VBAC) in women with diabetes. Multivariate analysis found that being treated for gestational diabetes was the most significant factor in determining caesarean section (OR 2.6.3 to 3.318 The study identified four groups: negative gestational diabetes (n = 2940). Multiple regression analysis showed that maternal nulliparity. Pregnancies in women with pre-existing diabetes were also more likely to be complicated by obstructed labour (P = 0.6 6. but caesarean section rates were 33.1.6% versus 20.2% in the women without diabetes.315 The proportion of women with diabetes in 1996 was 8.6% for the macrosomic group (RR 3. P < 0.9 to 2.0001). but this increased to 11.6).0001) were risk factors for having caesarean section.1.001) and caesarean section (29. [EL = 2−] A retrospective cohort study (n = 12 303) from the USA examined the relationship of diabetes and obesity on risk of caesarean section. Effect of diabetes on mode of birth A retrospective case–control study based on administrative data (n = 776 500) from Canada examined obstetric intervention and complication rates for women with and without pre-existing diabetes. 95% CI 0. untreated borderline gestational diabetes (n = 115) and known treated gestational diabetes (n = 143). untreated gestational diabetes was not a significant risk factor (OR 1. The study concluded that women with pre-existing diabetes needed close monitoring during pregnancy.316 The study found that diet-controlled gestational diabetes (P < 0.5 to 3.7% versus 13. [EL = 2+] A retrospective case–control study (n = 2924) undertaken in the USA compared the outcomes of macrosomic babies (4000 g or more) with those of babies weighing 3000–3999 g. multiple regression analysis showed that only pre-existing diabetes was an independent risk factor for having caesarean section. The study found that women with preexisting diabetes were significantly more likely to have caesarean section or induced labour than women without diabetes (P < 0. The study concluded that more research was needed to determine suitable interventions.42 per 1000 deliveries.90 per 1000 deliveries by 2001.7).7% in the untreated group and 13. [EL = 2+] 115 . 95% CI 1. the study noted marked increases in all these outcomes between 1996 and 2001 in women without diabetes. [EL = 3] A prospective cohort study (n = 166) from the USA examined route of birth of women with gestational diabetes. Nine studies were identified that investigated different methods of intervening in timing and mode of birth in women with diabetes.7%.02). Five studies were identified in relation to optimal timing of birth in women with diabetes.0001) and pre-existing diabetes (P < 0.

previous caesarean section. shoulder dystocia 1. [EL = 1+] A cohort study from Israel (n = 1542) examined the effect of intensive management of diet and three protocols for active elective management of route of birth on outcomes in women with gestational diabetes.0% versus 17. There were no differences between the groups at baseline. and those LGA but less than 4250 g being born vaginally after induction of labour. 95% CI 1.323 Those enrolled had gestational diabetes (n = 187) or pre-existing diabetes (n = 13).27 to 16. [EL = 3] A population-based cohort from the UK (n = 11791) examined antenatal risk factors associated with a woman having a caesarean section. the rate in the women without diabetes who had induced labour or caesarean section was 33.05). The study also found that babies of women with gestational diabetes who were equal to or greater than the 90 percentile in birthweight were more likely to have retarded lung development (P < 0. resuscitation of babies with positive pressure ventilation and low Apgar score (less than 3) at 1 minute and 5 minutes.4% versus 16. P = 0. induction of labour 11.9).8% (P < 0. the study did not examine any health professionalor healthcare-related factors. estimated fetal weight for caesarean section more than 4000 g.0%. estimated fetal weight for caesarean section more than 4500 g. seven had caesarean section and 44 had spontaneous labour.321 The study found that having gestational diabetes increased the risk of having a caesarean section (OR 2. n = 100).38 to 4. a number of other factors were also identified. In the active induction group 70 women had induction of labour.50. LGA 23. time of elective induction 38 weeks of gestation).8 per 1000 perinatal deaths in the group without gestational diabetes and 70. period C.8 mmol/litre. gestational age at birth and fetal presentation.02). period B.05).325 The study compared two time periods (before and after introduction of a protocol based on ultrasound estimates of fetal weight.9% versus 8.6% versus 21. Maternal risk factors were not examined and the study was undertaken in 1978 since when the neonatal survival rate has improved.0% versus 11. However. as did preexisting diabetes (OR 8. [EL = 2+] A cross-sectional study322 assessed routes of birth and pregnancy outcomes in 10 369 births in the USA.2%. The 116 . However.Diabetes in pregnancy A prospective cohort study from the USA (n = 53 518) examined the association between perinatal death and presence of gestational diabetes. The study concluded that active induction of labour at 38 weeks of gestation should be considered in women with insulinrequiring diabetes. including obstetric history and medical history. However.6%. The study showed that diabetes was associated with increased caesarean section. The results were as follows (period A versus period B versus period C): macrosomia (more than 4000 g) 17. eight had caesarean section and 22 had spontaneous labour. those more than 4250 g being born via caesarean section. with AGA fetuses being managed expectantly.0% versus 35. The study also examined biochemical markers in the placenta. neonatal head circumference. mean glucose less than 5.2% versus 0. [EL = 2+] A case–control study (n = 2604) undertaken in the USA examined the outcome of elective caesarean section due to macrosomia in women with diabetes.3 mmol/litre. mean glucose 5. There were significantly more LGA babies in the expectant management group compared with the active induction group (23% versus 10%. birthweight. estimated fetal weight for caesarean section more than 4000 g. [EL = 3] Regimens for inducing labour and their impact An RCT (n = 200) from the USA involving women with insulin-requiring diabetes compared the outcomes of active induced labour (accurate measurement of gestational development and induction of labour with intravenous oxytocin.92).320 The study found 33.5% versus 1.8 per 100 in the women without gestational diabetes who had spontaneous birth.3 per 100) in the group with gestational diabetes. outcome of last pregnancy.324 The results for the three periods of different protocols were compared (period A. umbilical cord and fetal membranes.2 per 1000 compared with 14. In the expectant management group 49 women had induction. caesarean section 20. The study concluded that intensive management of diet and active management of birth were beneficial to women with gestational diabetes and their babies. time of elective induction 40 weeks of gestation.60. parity.9% versus 14.7% (P < 0. Multiple regression showed that diabetes was a risk factor for caesarean section alongside maternal age. 95% CI 4. n = 100) with expectant management (close monitoring and insulin treatment.6% versus 18.005). There were three cases of ‘mild’ shoulder dystocia in the expectant management group and none in the active induction group.

The caesarean section rate was lower in women who had induction of labour than those who did not have induction of labour (OR 0.8 g versus 4132. wound separation (2 versus 0).326 Women with diabetes were more likely to have a caesarean section than women without diabetes (OR 2.Intrapartum care rate of shoulder dystocia was lower in macrosomic babies in the induced grouped compared with the non-induced group (7. During caesarean section the women with diabetes were more likely to have estimated blood loss above 1000 ml (P < 0.4% versus 18. 95% CI 1.04).8 years versus 29. 95% CI 1. the expectant management of pregnancy after 38 weeks of gestation did not reduce the incidence of caesarean section. analysis did not take into account baseline and surgical differences between groups. neonatal infection (1 versus 1).1% versus 21. blood transfusions (2 versus 0). [EL = 1+] A quasi-randomised study (n = 84) undertaken in the USA compared the outcomes of caesarean section (n = 44) and vaginal births (n = 40. In women with pre-existing diabetes.330 At baseline women with pre-existing diabetes were more likely to be obese (P < 0.18) or neonatal characteristics. 95% CI 1.23). P = 0.9).01). The study concluded that misoprostol was not beneficial for induction of labour in women with diabetes.329 The complications recorded for caesarean section versus vaginal birth were: morbidity (9 versus 0). OR 2.1).8%. However. [EL = 1+] 117 . 91 spontaneous vaginal births.8 g. There were five cases of shoulder dystocia in the induction group compared with six in the expectant management group.7). However. P < 0. but rather led to an increased prevalence of LGA babies (23% versus 10%) and shoulder dystocia (3% versus 0%).83 to 2.2) and wound complications (OR 3.01) and have a positive group B streptococcus status (P < 0.1. but if this is not pursued careful monitoring of fetal growth should be performed. neonatal hypoglycaemia (1 versus 0) and hyperchloraemic acidosis (1 versus 0). P = 0. but they were less likely to have meconium present (P = 0.7. total oxytocin dose (P = 0. 95% CI 1. 91 spontaneous vaginal births. The study recommended the use of ultrasound to estimate fetal weight and using this to determine method of birth. prematurity by weight (6 versus 5).8 to 7. 26 spontaneous and 14 induced) in women with gestational diabetes (3 hour 100 g OGTT).7. The study concluded that diabetes was a risk factor for wound complications after caesarean section. The study concluded that estimated fetal weight between 4000 g and 4500 g should not be considered an indication for inducing birth. [EL = 2+] A case–control study (n = 388) undertaken in the USA examined the risk of wound complication after caesarean section in women with pre-existing diabetes and women without diabetes.50 to 0. The rate of caesarean section was higher postprotocol compared with pre-protocol (25. 17 instrumental births and 26 caesarean sections. the study suggested that active induction of labour at 38 weeks of gestation should be considered in women with insulin-requiring diabetes. The study concluded that there was no advantage to preterm caesarean section in women with gestational diabetes. 95% CI 0.5 years.8 to 21. [EL = 2−] A retrospective cohort study using routinely collected data (n = 108 487) undertaken in the USA examined whether induction of labour increased caesarean section rates in women with diabetes.00. nor in the mean birthweight (4062. postpartum haemorrhage (P = 0.05) and to spend longer in the operating theatre (P = 0. expectant management group.01).01).2 to 6. the study was not explicitly conducted on women with diabetes.328 There was no difference between the groups at baseline.89). 18 instrumental births and 30 caesarean sections). [EL = 2+] An RCT (n = 273) undertaken in Israel compared induction of labour with expectant management in the presence of macrosomia (4000–4500 g).01).19).02).77. [EL = 1+] An RCT (n = 120) undertaken in the USA compared misoprostol with placebo for inducing birth in women with diabetes. The study found no difference between the groups during outpatient observation. Women with diabetes were more likely to have wound infection (OR 2.24). time from induction to birth (P = 0.327 At baseline the women in the induction group were significantly older than the expectant management group (30. [EL = 2−] Optimal timing of birth An RCT (n = 200) from the USA compared the outcomes of birth after 38 weeks of gestation in women with insulin-requiring diabetes. Given the risk associated with birth after 38 weeks of gestation.323 Those enrolled had gestational diabetes (n = 187) or pre-existing diabetes (n = 13). macrosomia more than 4000 g (5 versus 3). wound separation (OR 6.7%. There were no significant differences in the mode of birth between the groups (induction of labour group.

5 weeks. labour management and obstetric history.4% in the induction of labour group compared with 10. previous vaginal birth or VBAC (40% versus 17%. birthweight.001).324 The gestational age at birth for women with gestational diabetes was 39 ± 2.001). P > 0. induced labour. Logistic regression analysis showed that age. tobacco use. The study suggests that if pregnancy is not interrupted then the gestational age at birth is similar between women with diabetes and those without diabetes.05). sexually transmitted disease. After controlling for birthweight. Data on 423 women with diet-controlled gestational diabetes and 9437 women without diabetes were analysed. and neonatal outcomes do not differ between the two groups.20 for women with gestational diabetes). The VBAC group had more previous pregnancies (3. Apgar scores.318 The study identified four groups: negative gestational diabetes (n = 2940). P < 0.5 weeks and that of women without diabetes was 39 ± 1. chronic hypertension. P < 0.8 ± 1. false-positive gestational diabetes (n = 580).8 for women with borderline diabetes and 39. including gestational diabetes.05). The study recommended elective induction of labour for women with insulinrequiring diabetes in order to reduce the rate of shoulder dystocia. neonatal death or prolonged hospital stay after birth.05). fetal distress and breech birth. whilst diet-controlled gestational diabetes and chronic hypertension were not. shoulder dystocia.2% in the non-induced group who gave birth beyond 40 weeks of gestation (P < 0. pre-existing diabetes and gestational diabetes. [EL = 2+] A cohort study (n = 317) from Israel conducted between 1993 and 1995 examined the effect of intensive management of gestational diabetes with diet in relation to birth timing and outcomes and compared the effect with that for women without diabetes. cephalopelvic disproportion.9 weeks). P < 0. race.3 ± 1.2%. The rate of shoulder dystocia was 1. dysfunctional labour. The study found that gestational diabetes was not an independent risk factor for VBAC.8 weeks versus 39 ± 2. No differences in caesarean section rates or birthweights of babies were found. different ethnic mix (P < 0.332 The study found no significant difference between women with diabetes and the controls in gestational age at birth (38. The factors associated with unsuccessful VBAC were birthweight more than 4000 g. [EL = 2−] Vaginal birth after previous caesarean section A retrospective case-series (n = 10 110) from the USA examined the success of VBAC in women with previous obstetric complications.Diabetes in pregnancy A case–control study (n = 260) from Israel compared inducing labour at 38–39 weeks of gestation with allowing pregnancy to continue naturally in women with type 1 diabetes.001) and birthweight more than 4000 g (18% v 33%. augmented labour and previous vaginal birth were all predictors of successful VBAC. prolonged labour. There were no significant differences in gestational age at birth (39.001). [EL = 3] A retrospective cohort study (n = 25 079) from the USA examined whether women with dietcontrolled gestational diabetes were at increased risk of failed VBAC compared with women who did not have diabetes.4 versus 3.05). [EL = 2+] A case–control study (n = 428) from the USA examined the mean gestational ages at birth of babies of women with gestational diabetes and those in a control group without maternal diabetes. There were no differences among the groups in the rates of fetal distress or shoulder dystocia.1. [EL = 2+] 118 .8 ± 1. hypertension. and the proportion of babies weighing more than 4000 g was 18% compared with 13% (P < 0. 49% of the women with gestational diabetes and 67% of the women without diabetes attempted VBAC. The rate of shoulder dystocia was lower in the babies of women who had induction of labour at 38–39 weeks of gestation than in those without induction (1. different insurance profile (P < 0.8 weeks for women without diabetes.6. P < 0. hospital setting. 4 ± 2. untreated borderline gestational diabetes (n = 115) and known treated gestational diabetes (n = 143). white ethnic origin. P < 0. The success rate for attempted VBAC among women with gestational diabetes was 70% compared to 74% for women without diabetes.331 There were no differences between the two groups at baseline. induced labour. maternal age.333 Sixty-two percent of women who attempted VBAC were successful.334 The study involved 13 396 women who attempted VBAC.4% versus 10. 39. [EL = 2−] A case–control study (n = 3778) from Canada examined the relationship between gestational glucose intolerance (3 hour 100 g OGGT) and fetal outcomes. seen in university hospital (56% versus 42%.001).

006).76%. P = 0.8% compared with 7.Intrapartum care A retrospective cohort study (n = 428) conducted in the USA compared attempted VBAC in women with gestational diabetes to that in women without diabetes. The study found that those with previous gestational diabetes were more likely to have a future caesarean section (35.8%.332 One hundred and fifty-six women with gestational diabetes were matched with 272 controls.7. P = 0. 95% CI 1.2%). P < 0. large baby (3. general obstetric complication (14.9 g. [EL = 3–4] The preterm birth rate was 35.9%).4%). previous caesarean section (24. The birthweight in the gestational diabetes group was significantly greater than the control (3437. less likely to have a vaginal birth (64. This is thought to be an underestimate of the actual number of pregnancies that ended 119 .9% (261/3808) of pregnancies led to in utero losses (there were also two early neonatal deaths. The indications for elective and emergency caesarean section were presumed fetal compromise (28. Current practice The CEMACH enquiry reported that women with pre-existing diabetes had high rates of obstetric intervention with a 39% induction of labour rate compared with 21% in the general maternity population. [EL = 2+] No evidence was identified to suggest that the indications for caesarean section in preference to induction of labour are different for women with diabetes compared to women without diabetes.7%).001) and more likely to be white or Hispanic (P = 0. The parities were similar for the two groups. adjusted for maternal age and deprivation). although it is important to note that women who did not have a discussion also gave birth at an earlier gestational age. [EL = 3–4] The enquiry found that 6. P < 0. preterm rupture of membranes.9% versus 22. [EL = 3–4] The condition of the baby at birth was reported by the CEMACH enquiry: 2. The most frequent issues noted were poor management of maternal risks.5% versus 22.5%.001). The majority of women (65% of 382 women) were assessed as having optimal care during labour and birth and there was no association of sub-optimal care and pregnancy outcome.33 A lack of discussion was associated with poor pregnancy outcome (OR 4.4%).2%. large baby or polyhydramnios (8.9%). and the remainder were due to other clinical reasons.9%). failure to progress in labour (13. P = 0.2% versus 68. Additional case–control analysis showed an association with fetal or neonatal death. but the women with gestational diabetes were significantly older than the control group (P < 0. maternal request.540) but significant difference in failure of VBAC in the spontaneous labour group (18.4% were spontaneous preterm births (including preterm rupture of the membranes requiring induction) which is higher than in the general maternity population. but not with fetal congenital anomaly. The reasons given for induction of labour were that it was routine for women with diabetes (48. diabetes complications (2. but there were no significant differences in outcome Apgar scores or neonatal deaths. 21. lacerations or shoulder dystocia. presumed fetal compromise (9.5%) and diabetes complications (2.4%.4% were iatrogenic and 9.8 g versus 3191.9%. or unknown or inadequately described.7% versus 9. P < 0. Of the total births 26. The corresponding figure for the general maternity population is 0. inappropriate decisions relating to birth and inadequate fetal surveillance during labour or delay in acting on signs of fetal compromise. general obstetric complications (13.9% of which were for previous caesarean section.1%).001).9%). and the remainder were other clinical reasons. twins born live at 20 weeks of gestation who both died within 1 hour of birth).5%) and routine for diabetes (1.1% versus 77.4% in the general maternity population.2 to 12.20). There was no difference in failure of VBAC in the induced labour group (63. [EL = 3–4] The enquiry case–control study found that 8% (15/178) of women with poor pregnancy outcomes and 2% (4/202) of women with good pregnancy outcome had no details of discussion about timing and mode of birth in their medical records. The majority of iatrogenic preterm births were due to preterm caesarean sections.2 [EL = 3–4] The caesarean section rate was 67%.3%). which is three times higher than the general maternity population (24%). reason unknown or inadequately described. There were no differences for pre-eclampsia.001). maternal request.6% of live births had an Apgar score of less than 7 at 5 minutes.0. large baby.001) and more likely to have induction of labour (38. maternal request or routine for maternal diabetes.

7 per 1000 live births and still births (RR 4.4% at 32–36 weeks of gestation and 27. [EL = 3–4] The enquiry recommended that women should be involved in the decision–making process regarding timing and mode of birth. Arrangements should be in place to enable the rapid transfer of women with diabetes who choose not to deliver in a consultant-led obstetric unit. 95% CI 19.8% at 28–31 weeks of gestation. The NSF for diabetes20 states: [EL = 4] ‘Women with pre-existing diabetes: During labour.5% occurred at 24–27 weeks of gestation. and neonatal hypoglycaemia. 41.2 to 13.7 to 3. No neonatal deaths were reported after 42 weeks of gestation. 38.2 to 39. [EL = 3–4] Existing guidance The NICE induction of labour guideline12 recommended that women with pregnancies complicated by diabetes should be offered induction of labour before their estimated date for delivery. which occurs in approximately 24% of babies.5% at 32–36 weeks of gestation and 15. but other complications. [EL = 3–4] The perinatal death rate (the number of stillbirths and early neonatal deaths per 1000 live and stillbirths. The risk of macrosomia can be reduced by the achievement of near normal blood glucose levels during the third trimester.0).4 adjusted for maternal age) was significantly higher than the national rate of 8.4% at 37–41 weeks of gestation.5 per 1000 live births and stillbirths (RR 3.7 to 6. adjusted for maternal age) was significantly higher than the national rate of 5. 95% CI 5. 95% CI 1. Although the guideline reported that there were insufficient data clarifying the gestationspecific risk for unexplained stillbirth in pregnancies complicated by diabetes. 120 . hypoglycaemia and polycythaemia. are now rare. 95% CI 24.8.9). Continuous electronic fetal heart monitoring should be offered to all women with diabetes during labour and fetal blood sampling should be available if indicated. The induction of labour guideline is currently being updated and induction of labour for women with diabetes is excluded from the scope having been incorporated into the scope for the diabetes in pregnancy guideline.6. the GDG that developed the induction of labour guideline considered that it was usual practice in the UK to offer induction of labour to women with type 1 diabetes before 40 weeks of gestation. a midwife experienced in supporting women with diabetes through labour should be present and an appropriately trained obstetrician should be available at all times. including VBAC and provision of information for women. Jaundice is also more common. Macrosomia is associated with an increased risk of fetal injury and damage to the birth canal. 9. The woman’s blood glucose level should be maintained within the normal range in order to reduce the risk of neonatal hypoglycaemia. The stillbirth rate (26.3. The NICE intrapartum care guideline10 contains recommendations in relation to routine intrapartum care.8 to 33.7. 95% CI 3.6 per 1000 live births (RR 2. should difficulties arise.Diabetes in pregnancy before 24 weeks of gestation due to many women not accessing maternity services during the early stages of pregnancy. adjusted for maternal age) was significantly higher than the national rate of 3. Women who develop gestational diabetes: In the second and third trimesters of pregnancy. No stillbirths were reported after 42 weeks of gestation. 31.2% occurred at 24–27 weeks of gestation. 13. 95% CI 3.6% at 37–41 weeks of gestation.7).8 per 1000 live births and stillbirths.7% at 28–31 weeks of gestation. [EL = 3–4] Of the pregnancies continuing at 24 weeks of gestation 87/3474 ended in stillbirth and 3449 live births.0 to 4. the main risks to the baby are macrosomia (excessive fetal growth). Of the stillbirths 17. Problems during the neonatal period can be reduced by the achievement of tight blood glucose control during labour. such as respiratory distress syndrome.3.8. Of the neonatal deaths 38. 7. [EL = 3–4] The neonatal death rate (number of neonatal deaths per 1000 live births.8. affecting approximately 30% of the offspring of women with type 1 diabetes. The NICE caesarean section guideline13 contains recommendations in relation to caesarean section. including electronic fetal monitoring.

Intrapartum care Improving pregnancy outcomes: The rate of shoulder dystocia can be decreased by the use of ultrasound monitoring and elective delivery of those babies weighing over 4250 g.’ Evidence statement Eight studies examined the epidemiology of women with diabetes requiring intervention in the timing and mode of birth. The studies showed that stillbirth and shoulder dystocia were the greatest risks involved. Caesarean section rates were lower in women who were actively managed for induction of labour. tight blood glucose control during labour can reduce the risk of neonatal hypoglycaemia and hence reduce the need for admission to a neonatal intensive care unit. However. it should be remembered that some women’s experience of a “medicalised” and high-intervention labour and delivery is a negative and frightening one. but diabetes was not a complete contraindication. Close monitoring and prompt intervention can improve outcomes for both the mother and her baby. therefore. Routine induction of labour for women with diabetes at 38–39 weeks of gestation reduces the risk of stillbirth and shoulder dystocia without increasing the risk of caesarean section. induction of labour and caesarean section. but these none of these studies was specifically designed to address the optimal timing of birth in women with diabetes. there was insufficient evidence to determine the precise gestational age at which elective induction of labour should be offered. caesarean section rates remain high even when diabetes is controlled and macrosomia is not present. For example. The remaining studies allowed comparison of gestational ages at birth between babies of women with diabetes and those of women without diabetes. Nine studies investigated different methods of intervening in the timing and mode of birth. Preparation for surgical birth (caesarean section) is considered in Section 6. Three studies examined VBAC and showed that birthweight was the main factor in determining success.2. Routine induction of labour at 38–39 weeks of gestation did not increase the rate of caesarean section. Five studies were considered in relation to optimal timing of birth in women with diabetes. and if they are supported by calm and competent professionals. There were more LGA babies and cases of shoulder dystocia in the expectant management groups. the majority of the studies were conducted in the USA where sociocultural and health service factors are different to the UK. However. In the absence of evidence to determine whether 121 . Diabetic pregnancy is associated with an increased risk of complications during labour and delivery. The GDG’s discussions highlighted the need to balance the risk of fetal lung immaturity which may be associated with induction at 36–37 weeks of gestation against the risk of stillbirth associated with later induction. An RCT involving women with insulin-requiring diabetes and a case–control study involving women with type 1 diabetes compared elective induction of labour at 38–39 weeks of gestation with expectant management. There is evidence that fetal weight and risk of stillbirth are the determining factors in whether women with diabetes undergo caesarean section. and caesarean section was associated with higher levels of complications and adverse outcomes than vaginal birth. From evidence to recommendations Evidence shows that women with diabetes are more likely to undergo induction of labour and/or caesarean section than women without diabetes. For example. In one study. are involved in decision-making and kept informed. which are associated with fetal macrosomia. Diabetes during pregnancy and macrosomia were the greatest risk factors affecting outcome of birth. ultrasound was found to determine accurately the presence or absence of macrosomia in 87% of women scanned. evidence shows that factors related to the behaviour of healthcare professionals and the organisation of health service impact on outcome. No evidence was identified to suggest that induction of labour should be conducted differently in women with diabetes compared to other women (including oxytocin protocols and electronic fetal monitoring). Women with diabetes were more likely to have unsuccessful VBAC compared to women without diabetes. Healthcare professionals should. This need not be the case if they are helped to feel in control. The reasons for intervention in the mode of birth in women with diabetes are to prevent stillbirth and shoulder dystocia. However. inform women with fetal macrosomia of the risks and benefits of vaginal birth. However.

6.05. 8. or by elective caesarean section if indicated. be considered carefully in terms of the effect on the woman and the baby. [EL = 1+] A narrative non-systematic review reported that pain and/or stress following surgery and trauma impairs insulin sensitivity by affecting non-oxidative glucose metabolism. 8.2 Analgesia and anaesthesia Description of the evidence Labour and birth can be stressful for any woman.Diabetes in pregnancy elective birth through induction of labour. prevention of metabolic disturbances (abnormal acid–base status leading to ketoacidosis) and haemodynamic control (with an emphasis on prevention of hypotension). Although this was a small study with only eight people without diabetes in each treatment group. In women with diabetes these stresses may make diabetes more difficult to control. induction of labour and caesarean section. No evidence was identified to suggest that the indications for elective caesarean section in preference to induction of labour in women with diabetes would be any different to those in women without diabetes. 122 . which may complicate obstetric analgesia and anaesthesia.5 ± 1. Recommendations for timing and mode of birth Pregnant women with diabetes who have a normally grown fetus should be offered elective birth through induction of labour. an RCT compared epidural anaesthesia plus general anaesthesia with general anaesthesia alone in people undergoing colorectal surgery for non-metastatic carcinoma. Diabetes should not in itself be considered a contraindication to attempting vaginal birth after a previous caesarean section.337 While these observations were not drawn from labouring women with diabetes the suggestion is that glucose regulation can be improved with administration of analgesia (pain relief) in stressful states. Pregnant women with diabetes who have an ultrasound-diagnosed macrosomic fetus should be informed of the risks and benefits of vaginal birth. resulting in otherwise preventable morbidities and even mortality. Relevant factors to consider in terms of women with diabetes are glycaemic control.4 micromol/kg/ min. Any interventions during labour and birth should. [EL = 4] Acid–base status One cohort study and one case–control study investigated acid–base status (and neonatal Apgar scores) in women with diabetes undergoing elective caesarean section with epidural or spinal anaesthesia. Evidence shows that diabetes should not be considered a contraindication to attempting VBAC. However.2 micromol/kg/min. P < 0.05). or elective caesarean section if indicated.7 ± 1. postoperative glucose production.5 ± 1.336 Epidural anaesthesia plus general anaesthesia reduced the perioperative increase in blood glucose compared with general anaesthesia alone (intra-operative glucose production.335 Additional factors to consider are comorbidities such as neuropathy and obesity.8 micromol/kg/min versus 10. There were no research recommendations relating to the timing and mode of birth for women with diabetes or for the implications for future pregnancies and births. after 38 completed weeks. the GDG’s view was that elective birth should be offered after 38 completed weeks of gestation. should be offered before 38 weeks of gestation. Glycaemic control No clinical studies were identified in relation to the effects of analgesia or anaesthesia on perioperative glycaemic control in women with diabetes. therefore.2 ± 1.9 micromol/kg/min versus 10. it suggests that there may be a benefit in regional anaesthesia for women with diabetes facing caesarean section compared with general anaesthesia. P < 0.

30 (SE 0. However. be preferred for women with diabetes. therefore.6%) and unplanned dural puncture (0. traumatic needle placement (evidence of blood.339 The study involved 567 people with preexisting peripheral sensorimotor neuropathy or diabetic polyneuropathy who were investigated for neurological injury after neuraxial blockade.01) and 7.1% to 1. However. respectively. [EL = 2+] The case–control study. Two people (0. [EL = 2+] Haemodynamic control No clinical studies were identified in relation to the effects of analgesia or anaesthesia on perioperative haemodynamic control in women with diabetes. women with diabetes and their babies had a 25–50% reduction in pyruvate concentrations in maternal venous blood and neonatal umbilical venous and arterial blood compared to women without diabetes and their babies (P = 0.335 All the women were scheduled for elective primary or repeat caesarean section using spinal anaesthesia at term. respectively. One risk was that women with diabetes tend to have a higher resting gastric volume (slower gastric emptying) than women without diabetes. increasing the risk of Mendelson syndrome. 95% CI 0. a narrative non-systematic review reported that obesity is a risk factor for obstetric analgesia and anaesthesia. and hypotension was prevented in all women.338 At birth there were no significant differences between women with diabetes and those without diabetes in terms of arterial blood acid–base status.27 (SE 0. Dextrose-free intravenous solutions were used for volume expansion before induction of anaesthesia.3%) experienced new or progressive postoperative neurological deficits and 65 technical complications occurred in 63 people (11. respectively. Mean umbilical artery pH in women with diabetes and women without diabetes were 7. which was undertaken in the 1980s.4%.Intrapartum care The cohort study assessed whether epidural anaesthesia in women with diabetes undergoing elective caesarean section was associated with abnormal acid–base and glucose status in the woman and baby compared with epidural anaesthesia in women without diabetes.42 (SE 0.01). nor between babies of women with diabetes and babies in the control group.340 [EL = 4] No further clinical studies were identified in relation to factors affecting the choice of analgesia or anaesthesia in women with diabetes. There were no significant differences in acid–base values between women with diabetes and those without diabetes. The study suggests that epidural anaesthesia in women with diabetes is associated with normal acid–base status in the mother and baby. with one baby of a woman with diabetes having a score of less than 7 at 1 minute and the remaining 19 babies in both groups having scores of more than 7 at 1 and 5 minutes.01). The most common complications were unintentional elicitation of paraesthesia (7.01). no clinical data were provided to support either of these statements. However. Mean maternal artery pH in women with diabetes and women without diabetes were 7. They also suggested that sedatives. [EL = 2+] Obesity No clinical studies were identified in relation to the effects of coexisting obesity on analgesia or anaesthesia in women with diabetes.35 (SE 0.335 Another risk was that irreversible brain damage could occur if hypoglycaemia was allowed to develop 123 .33 (SE 0. Mean umbilical vein pH in women with diabetes and women without diabetes were 7. which results from aspiration of gastric contents into the lungs following vomiting or regurgitation in obstetrical patients. Apgar scores were similar in both groups. The study concluded that the risk of severe postoperative neurological injury is high in the population of people with pre-existing neuropathy and healthcare professionals should be aware of this when developing and implementing regional anaesthetic care plans. narcotic analgesics and muscle relaxants are similarly of benefit to women with diabetes. There were no infectious or haematological complications.6%).9%).40 (standard error (SE) 0.001). However. 1.01) and 7. a narrative non-systematic review suggested that increased risks were associated with general anaesthesia. involved ten women with rigidly controlled type 1 diabetes and ten healthy women without diabetes.006) and 7. The authors of the study suggested that anaesthetics such as nitrous oxide and intravenous agents such as thiopentone are virtually free of metabolic effects and may. Neuropathy A retrospective cohort study was identified in relation to the effects of coexisting neuropathy on analgesia and anaesthesia in women with diabetes. nor in terms of neonatal umbilical venous or arterial blood acid–base status or neonatal Apgar scores.1%).

are satisfactory. However. In particular it was reported that regional anaesthesia can accentuate haemodynamic distortions in women with diabetes in the presence of polyhydramnios and increased segmental spread of the nerve blockade. provided that fluids. the GDG’s discussion included consideration of the possibility of prolonged labour with epidural anaesthesia increasing the risk of DKA. and that this should normally be achieved through adjustment of intravenous insulin delivery in response to regular blood glucose testing.Diabetes in pregnancy during general anaesthesia and surgery. However. and so the GDG has not made any recommendations specific to epidural or spinal anaesthesia. Such protocols should ensure that near-normoglycaemia is maintained without the risk of acute decompensation. No clinical studies were identified in relation to the effects of coexisting obesity on analgesia and anaesthesia in women with diabetes. [EL = 4] The review also reported that regional anaesthesia (as used for caesarean section). A narrative non-systematic review suggested that glucose regulation could be improved with administration of analgesia in stressful states. Evidence shows that the presence of (symptomatic) autonomic neuropathy in women with diabetes is a risk factor for obstetric anaesthesia. [EL = 4] Existing guidance The NICE guideline for the management of type 1 diabetes recommends that hospitals ensure that protocols for inpatient procedures and surgical operations are in place and used. A cohort study showed that neuraxial blockade carries an increased risk of severe postoperative neurological injury in people with pre-existing neuropathy. etc. and that irreversible brain damage could occur if hypoglycaemia was allowed to develop during general anaesthesia and surgery. which includes epidural anaesthesia and spinal anaesthesia. For labour analgesia. It was also suggested that a delay in returning to consciousness following general anaesthesia could prolong the time before routine metabolic management with insulin could be reinstituted.. Long duration of diabetes. The GDG’s view is that diabetes is not in itself a contraindication to restrict the duration of labour. dose or provision of fluids) in women with diabetes. A small RCT in people undergoing colorectal surgery showed that epidural anaesthesia plus general anaesthesia reduced the perioperative increase in blood glucose compared with general anaesthesia alone. hyperglycaemia and any opioid are thought to slow gastric emptying and increase risks for anaesthesia. Although no clinical evidence was found specifically in relation to coexisting diabetes and obesity as risk factors for 124 . The review also noted the possibility of women with diabetes inadvertently being given glucose orally rather than intravenously. Two observational studies showed that epidural anaesthesia and spinal anaesthesia are associated with normal acid–base status in women with diabetes undergoing elective caesarean section and with normal acid–base status and Apgar scores in their babies. No clinical studies were identified in relation to the effects of analgesia and anaesthesia on perioperative haemodynamic control in women with diabetes.335 Vomiting is a consequence of hypotension caused by sympathetic nerve blockade and it may exacerbate metabolic disturbances. carries risks for women with diabetes. blood glucose concentrations. The presence of DKA would alter the management of the need for urgent birth. No evidence was identified to suggest that epidural or spinal anaesthesia should be used any differently (in terms of monitoring. an epidural nerve blockade may reduce metabolic expenditure and avoid the emetic effects of opioids. Evidence statement No clinical studies were identified in relation to the effects of analgesia or anaesthesia on perioperative glycaemic control in women with diabetes. a narrative non-systematic review highlighted that obesity is in itself a risk factor for analgesia and anaesthesia. From evidence to recommendations Evidence suggests that epidural anaesthesia and spinal anaesthesia are associated with normal acid–base status in women with diabetes.

important that anaesthetists have access to a complete medical history including information about the extent of neuropathy (particularly autonomic neuropathy) and obesity to assess fitness for anaesthesia and that monitoring begins before anaesthesia is initiated. Some fetuses develop a pattern of hyperinsulinaemia to cope with the regular excessive glucose transfer across the placenta where the maternal diabetes is poorly controlled122 (see Section 4. Other babies who have not developed hyperinsulinaemia in fetal life may respond to maternal hyperglycaemia in labour with sufficient insulin production in the 1–2 hours following birth to cause transient hypoglycaemia. Research recommendations for analgesia and anaesthesia What are the risks and benefits associated with analgesia and anaesthesia in women with diabetes? Why this is important The increasing number of women with diabetes and the high rate of intervention during birth emphasise the need for clinical studies to determine the most effective methods for analgesia and anaesthesia in this group of women. Very few women are likely to undergo general anaesthesia during labour and birth. one study included women with type 1 and type 2 diabetes and one study involved only women with gestational diabetes. It is. Recommendations for analgesia and anaesthesia Women with diabetes and comorbidities such as obesity or autonomic neuropathy should be offered an anaesthetic assessment in the third trimester of pregnancy. in the absence of adequate glucose intake. The research studies should investigate the effect of analgesia during labour. obesity alone has been reported to increase problems with intubation. general anaesthesia carries a risk of hypoglycaemia and women recovering from general anaesthesia will lack hypoglycaemia awareness. and it should continue after the birth until the woman is fully conscious. will lead to severe and prolonged hypoglycaemia. All six studies found maternal hyperglycaemia during labour to be associated with neonatal hypoglycaemia. 6. However. Regular blood glucose monitoring during anaesthesia is. and the cardiovascular effects of spinal anaesthesia and vasopressors on diabetic control. therefore. therefore. Eight observational studies were identified that considered the effect of maternal blood glucose control during labour and birth on neonatal blood glucose levels.3 Glycaemic control during labour and birth Description of the evidence Neonatal hypoglycaemia Neonatal hypoglycaemia can occur for two reasons. If general anaesthesia is used for the birth in women with diabetes. Four studies involved only women with type 1 diabetes. risk of thromboembolism.Intrapartum care obstetric anaesthesia. recommended for women with diabetes: monitoring should occur at 30 minute intervals to prevent hypoglycaemia remaining undetected and/or untreated for longer periods. blood glucose should be monitored regularly (every 30 minutes) from induction of general anaesthesia until after the baby is born and the woman is fully conscious. In contrast the term baby of a woman without diabetes demonstrates rather sluggish insulin responses to glycaemic stimuli and shows a tendency to relatively high blood glucose levels after feeding in the newborn period.1). 125 . The woman’s need for insulin will decrease significantly after the birth (see Section 8. As newborns these babies have a persisting autonomous insulin secretion which. stress response with pre-eclampsia and risk of post-epidural neurological problems. with some overlap in individual cases.1) and this can also lead to hypoglycaemia.

0 ± 1. then decrease by 1 U/hour to a minimum of 0.0–7.5 mmol/litre.0 mmol/litre) occurred in 11 babies.0003). 100 ml/hour via IMED® pump hourly blood glucose estimation by glucose meter insulin infusion by intravenous pump mounted onto the intravenous line. The maternal blood glucose concentration at birth correlated positively with the neonatal blood glucose concentration at birth (r = 0.7 mmol/litre were treated using intravenous glucose. Boluses of 2–5 units of regular insulin were given additionally if the plasma glucose level exceeded 5. The degree of hypoglycaemia was mild and rarely required admission to intensive care or intravenous treatment. Blood glucose was maintained at 6.001) and negatively with the neonatal blood glucose concentration 2 hours after birth (r = −0.3 ± 2.3–5 mmol/ litre.5% (38 babies).0 mmol/litre – refer to medical record or contact diabetes team for advice about subcutaneous insulin dose before next main meal – stop intravenous fluids and insulin 30 minutes after subcutaneous insulin.0–7. Blood glucose at birth was 6. Babies with blood glucose less than 2.Diabetes in pregnancy A retrospective study of 53 babies born to women with type 1 diabetes342 measured plasma glucose concentrations at birth and 2 hours later. P < 0.5 U/hour – if more than 7. 5% glucose) at a rate of 125 ml/hour.9% saline.6 mmol/litre. 156 with type 2 diabetes). [EL = 2++] A standardised intravenous protocol for insulin and dextrose therapy in labour and birth was assessed in 25 women with insulin-treated diabetes. [EL = 2+] A prospective study included 122 pregnancies in 100 women with type 1 diabetes.0 mmol/litre and not rising. Thirty-seven percent (11/30) of babies born to women whose blood glucose concentration at birth was 7. There was only one case of maternal hypoglycaemia. Introduction • • • • 126 .7 mmol or less). Forty-seven percent (36/76) of babies born to women who had CBG concentrations above 5 mmol/litre before birth developed neonatal hypoglycaemia (less than 1.1 mmol/litre.7 mmol/litre) compared with 14% (6/42) of babies born to women with CBG concentrations less than 5 mmol/litre (P = 0.7 mmol/litre received enteral feeds or intravenous glucose as dictated by their clinical condition.0 mmol/litre according to glucose meter: – if less than 4.82. Neonatal blood glucose correlated with maternal blood glucose at birth (r = −0. The protocol was as follows: nil by mouth until after the birth of the baby start intravenous dextrose 10% in 500 ml.46. P < 0. 0.345 Adjustments to insulin infusion rates were determined by trends in blood glucose as well as by absolute concentration. The mean intrapartum blood glucose level was significantly lower in mothers of babies without hypoglycaemia (P < 0. to a minimum of 0. P < 0. The incidence of neonatal hypoglycaemia (plasma glucose concentration less than 1. No babies born to women whose blood glucose concentration was less than 7. initially at 2 U/ hour when blood glucose more than 7 mmol/litre (50 U human soluble insulin in 50 ml. then increase by 0.05).8 mmol/litre.0 mmol/litre and not falling.8 mmol/litre for a mean of 6 hours (range 1–29 hours) before birth.001).7 mmol/litre) was 16.344 On the day of birth all women received an intravenous infusion of 10% invert sugar (5% fructose. Babies with blood glucose concentrations less than 1.5 U/hour • after delivery of the placenta: – halve the rate of insulin infusion.5 U/hour – adjust as before to maintain blood glucose 4.1 mmol/litre developed hypoglycaemia.1 mmol/litre or more developed hypoglycaemia (plasma glucose 1. Neonatal hypoglycaemia (plasma glucose less than 2. The authors reported that the best results were achieved when the desired glucose control was maintained for at least 8 hours before birth.58.9–5. The rate was adjusted if plasma glucose level was less than 2. Babies with plasma glucose concentration less than 1. Intravenous insulin was administered with an infusion pump at a rate of 1–4 U/hour to maintain plasma glucose concentration at 3.01). [EL = 2++] A prospective study included 233 women with insulin-requiring diabetes (77 with type 1 diabetes. 2 ml/hour) • adjust insulin infusion rate to maintain blood glucose 4.343 Intravenous glucose and/or insulin was infused during labour to maintain CBG concentrations at 3.0 mmol/litre were all treated routinely with intravenous glucose for 3–24 hours with none showing symptoms of hypoglycaemia.

P < 0.05). mean HbA1c throughout pregnancy (r = 0. P < 0. maternal age.3 ± 0. Mean CBG during labour was 4.2 mmol/litre in 19 babies. [EL = 2++] A study compared CSII (28 women with type 1 diabetes) with constant intravenous insulin infusion (37 women with type 1 diabetes). intravenous insulin infusion by syringe pump adjusted according to hourly CBG measurements.5 mmol/litre. neonatal blood glucose was less than 2.2 mmol/ litre. P < 0. Five babies developed hypoglycaemia. neonatal blood glucose was measured as less than 2.2 mmol/litre.2 ± 1. If the maternal blood glucose was above 9. After logistic regression (HbA1c in third trimester. In period 1.025). Mean blood glucose in women treated using constant intravenous insulin infusion was 7.0001).7 ± 1. In the 127 .3 g/hr). in all except two women. When maternal blood glucose stayed within the target of 4.6 mmol/litre (range 3.3.025).8 versus 2. [EL = 2+] An observational study of 107 consecutive singleton pregnancies in women with type 1 diabetes349 measured maternal HbA1c throughout pregnancy.346 Mean blood glucose during labour in women treated using CSII was 4.8 mmol/litre in all other women (P = 0. In the constant intravenous insulin infusion group there were eight cases of neonatal hypoglycaemia (less than 1.8 ± 0.7 mmol/litre). The target CBG range for metabolic control was 2.28).9 mmol/litre (ideally 3. and urine testing for ketone bodies.3% of women CBG was within the desired range. P < 0. Babies with hypoglycaemia were treated using intravenous glucose.3–6. There was no relationship between neonatal blood glucose and HbA1c throughout the third trimester (r = −0.10) or HbA1c at booking (r = 0. in period 2.8).33.5 ± 1.1 mmol/litre).5 for all others. Maximum maternal blood glucose during labour was higher in babies with perinatal asphyxia than in those without (9. There was a significant negative correlation between neonatal blood glucose and mean maternal blood glucose in labour (r = −0. Both studies found maternal hyperglycaemia during labour to be associated with fetal distress. Perinatal asphyxia was also associated with gestational age and development of vasculopathy during pregnancy but not with vasculopathy before pregnancy. When maternal blood glucose was greater than 8. Mean blood glucose in women treated using constant intravenous insulin infusion was 7. whereas in the CSII group there were no cases of neonatal hypoglycaemia (P < 0.8–5.8). [EL = 2++] A prospective study of 85 women with gestational diabetes347 (54 insulin-treated) was undertaken with the aim of assessing a standardised protocol for maintaining glycaemic control during labour and the effect of maternal glycaemic control during labour on neonatal hypoglycaemia (two or more glucose values less than 1. the neonatal blood glucose was always low (1. with intravenous glucose used in 14).6– 8.0–8.1 mmol/litre (range 5.346 Mean blood glucose during labour in women treated using CSII was 4.02). SGA.7 ± 3. When maternal blood glucose was over 10 mmol/litre. maternal blood glucose throughout labour and birth and neonatal blood glucose.0 mmol/litre there was no relationship with neonatal blood glucose. duration of diabetes or White’s classification of diabetes.8 mmol/litre in the group with neonatal blood glucose levels less than 2. compared with 4.5 mmol/litre.0 ± 3.05). A prospective study of 149 babies of women with type 1 diabetes350 found perinatal asphyxia in 27% (40).9 ± 2. insulin treatment) maternal blood glucose in the last 2 hours of labour was associated with neonatal hypoglycaemia (P < 0.7 versus 7.8–6.6–8.6 mmol/litre (range 3.3. [EL = 2++] A study involving women with type 1 diabetes compared CSII (insulin pump therapy) (n = 28) with constant intravenous insulin infusion (n = 37).2 ± 1. [EL = 3] Fetal distress Two observational studies were identified that considered the effect of maternal blood glucose control during labour and birth on fetal distress.001). neonatal hypoglycaemia was recorded in seven babies (less than 2.0 mmol/litre neonatal blood glucose was always less than 2.8–5. P < 0.05).01).5 ± 3.7 mmol/litre).0 mmol/litre.Intrapartum care of the protocol was associated with a decrease in the incidence of hypoglycaemia in babies born to women with diabetes from 68% to 39% (P < 0.0.1 mmol/litre and in 82. with intravenous glucose used in four). [EL = 2+] An observational study348 compared the effect of a policy change following an intensive effort to improve pre-pregnancy care and advice with a relaxation of targets for blood glucose control during labour. Mean maternal blood glucose at birth was 7. The protocol consisted of: intravenous glucose (8.1 mmol/litre (range 5.11).8 ± 0. preterm birth.

Two studies found maternal hyperglycaemia during labour to be associated with fetal distress. [EL = 1+] A non-randomised study compared CSII (28 women with type 1 diabetes) with constant intravenous insulin infusion (37 women with type 1 diabetes).8). Neonatal hypoglycaemia (blood glucose less than 0. hypoglycaemia due to clinical practice. the lack of randomisation may have meant that women who used CSII during labour and birth were self-selected as those who had better glycaemic control.351 There was no difference in mean intrapartum maternal CBG levels in the rotating fluids and intravenous insulin groups (5. [EL = 2++] Controlling glycaemia during labour and birth An RCT investigated whether continuous intravenous insulin infusion provided a greater degree of intrapartum maternal glycaemic control in women with gestational diabetes than rotating between glucose-containing and glucose-free intravenous fluids.20 Evidence statement Eight studies were found that showed that neonatal hypoglycaemia is more likely to occur in babies of women with high blood glucose concentration during labour and birth.28).025). An RCT found continuous intravenous insulin infusion and rotating between glucose-containing and glucose-free intravenous fluids resulted in similar maternal blood glucose levels in women with insulin-requiring gestational diabetes during labour and birth.346 Mean blood glucose during labour in women treated using CSII was 4. suboptimal blood glucose monitoring. P < 0. and hyperbilirubinaemia were also similar between the two treatments.Diabetes in pregnancy constant intravenous insulin infusion group there was acute fetal distress in 27% of cases and a caesarean section rate of 38%. continuous intravenous insulin infusion and a rotation of intravenous fluids between glucosecontaining and glucose-free fluids achieve similar intrapartum glycaemic control.6–8. Current practice The CEMACH enquiry found sub-optimal glycaemic control during labour and/or birth was not associated with poor pregnancy outcome in women with pre-existing diabetes. the lack 128 . In the CSII group there was fetal distress in 14. The enquiry panels identified cases of inappropriate intravenous insulin/dextrose regimen. A non-randomised comparative study found that CSII was associated with better glycaemic control in women with type 1 diabetes during labour and birth than was intravenous insulin infusion. Apgar scores at 1 minute and 5 minutes. P = 0.8 ± 0. However.2 ± 1. However.33 [EL = 3–4] The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that 47% of the women with type 1 diabetes and 41% of the women with type 2 diabetes had suboptimal glycaemic control during labour and birth (P = 0.33 [EL = 3–4] Existing guidance The NSF for diabetes recommends tight glucose control during labour to reduce the risk of neonatal hypoglycaemia.1 mmol/litre (range 5.3% of cases (P < 0.001).6 mmol/litre within the first 24 hours was found to be 6.8–5. respiratory distress.7% in the rotating group and 19% in the continuous intravenous insulin infusion group.3.77 ± 0. admission to the NICU. delay in starting intravenous insulin/dextrose regimen.05).48 mmol/litre versus 5.99 mmol/litre. The study suggests that in women with insulin-requiring gestational diabetes. poor management of hypoglycaemia and other clinical practice issues in both groups of women. Among women with poor pregnancy outcome 49% (79/162) were documented as having sub-optimal control in the first trimester compared with 48% (97/202) of the women with good pregnancy outcome. Birthweight. Ten percent of the women with type 1 diabetes and 29% of the women with type 2 diabetes were not given intravenous insulin and dextrose during labour and/or birth (P < 0.73 ± 0. Mean blood glucose in women treated using constant intravenous insulin infusion was 7. shoulder dystocia.001) and a caesarean section rate of 25% (P < 0.6 mmol/litre (range 3. Maintaining maternal blood glucose in the range 4–7 mmol/litre during labour and birth reduces the incidence of neonatal hypoglycaemia and reduces fetal distress. poor management of sliding scale. but the difference was not statistically significant.89).

no randomised controlled trials have compared the effectiveness of intermittent subcutaneous insulin injections and/or CSII with that of intravenous dextrose plus insulin during labour and birth. 129 .Intrapartum care of randomisation in the study means that self-selection of CSII by women who were better at controlling their blood glucose cannot be ruled out. Intravenous dextrose and insulin infusion is recommended during labour and birth for women with diabetes whose blood glucose is not maintained at between 4 and 7 mmol/litre. However. Randomised controlled trials are therefore needed to evaluate the safety of intermittent insulin injections and/or CSII during labour and birth compared with that of intravenous dextrose plus insulin. In formulating their recommendations. neonatal hypoglycaemia and respiratory distress). no clinical studies were identified that evaluated the optimal method of maintaining glycaemic control during labour and birth. In the absence of such evidence the GDG’s consensus view was that intravenous dextrose and insulin infusion should be considered for women with type 1 diabetes from the onset of established labour. However. and the practicalities such as permitting greater mobility. This leaves the possibility for women who are able to maintain their blood glucose in the range 4–7 mmol/litre using MDI insulin injections or CSII to experience labour and birth without having intravenous insulin regimens. Women with type 1 diabetes should be considered for intravenous dextrose and insulin infusion from the onset of established labour. capillary blood glucose should be monitored on an hourly basis in women with diabetes and maintained at between 4 and 7 mmol/litre. The GDG noted that these options may be associated with greater maternal satisfaction because of the psychological benefits of allowing women to take control of their diabetes during labour and birth. and that intravenous dextrose and insulin infusion is recommended for women with diabetes whose blood glucose is not maintained between 4–7 mmol/litre. Recommendations for glycaemic control during labour and birth During labour and birth. The potential benefits of intermittent insulin injections and/or CSII over intravenous dextrose plus insulin during the intrapartum period include patient preference due to the psychological effect of the woman feeling in control of her diabetes and having increased mobility. the GDG placed a high value on recommending that blood glucose be maintained in this range during labour and birth without being prescriptive about how it is maintained. Research recommendations for glycaemic control during labour and birth What is the optimal method for controlling glycaemia during labour and birth? Why this is important Epidemiological studies have shown that poor glycaemic control during labour and birth is associated with adverse neonatal outcomes (in particular. From evidence to recommendations Evidence shows that maintaining blood glucose in the range 4–7 mmol/litre during labour and birth reduces the incidence of neonatal hypoglycaemia and fetal distress.

0025). Hypoxia causes polycythaemia and hyperviscosity. hyaline membrane disease. Diabetes was associated with increased caesarean section rates.0002) and respiratory distress (12% versus 1%. The women with diabetes used insulin by infusion pump or split-dose therapy. complication necessitating admission to a NICU. The incidence of respiratory distress syndrome and mortality was assessed in a cohort study356 involving 23 babies selected from a total of 30 babies born to women with diabetes who developed hypoglycaemia after birth.353 Information on incidence and presentation of thrombosis is very limited and has been gathered mainly from case reports and case series. P = 0.0002).003). or respiratory and metabolic support). Nearly all the low blood glucose determinations (less than 1.354 it is still important that treatment is provided promptly for those requiring it (e.g. seven babies treated with long-acting epinephrine plus intravenous glucose.g. babies of women with diabetes were 130 . hypoglycaemia (14% versus 1%. which is the period of greatest risk of low blood glucose occurring in babies born to women with diabetes.008).7% of babies of women with diabetes and 1. maternal factors resulting in IUGR.352 Thrombosis is a rare. symptoms of hypoglycaemia and interventions provided were assessed. and four babies treated with long-acting epinephrine only. Timing of blood glucose levels. [EL = 2+] A case–control study357 investigated factors that contribute to neonatal hypoglycaemia in babies of women with diabetes. pre-eclampsia (15% versus 5%. hyperbilirubinaemia (46% versus 23%. None of the 66 babies investigated developed symptomatic hypoglycaemia or required intravenous glucose. [EL = 2+] A cross-sectional study322 assessed routes of delivery and pregnancy outcomes in 10 369 births in the USA.05). These include fetal macrosomia.3% of controls (P = 0. P = 0. Complications of babies born to women with diabetes included LGA (41% versus 16%. P = 0. hypoglycaemia.7 7.035) and caesarean section (55% versus 27%. for which babies may require surfactant. cardiomyopathy. infant respiratory distress syndrome. Immediately after birth. transient tachypnoea normalises within 3 days of birth). There were no significant differences in incidence of respiratory distress syndrome or mortality rates between the three groups. Risk factors include maternal diabetes. P = 0. hypocalcaemia. polycythaemia and the use of intravascular catheters in preterm babies. hypomagnesaemia. with the goal of normalising fasting blood sugars and HbA1c. These babies were divided into the following three groups: 12 babies treated with intravenous glucose. P = 0. Apgar scores and mortality rates were similar for the two groups. polycythaemia and hyperviscosity. While some of the morbidities listed above correct themselves within a period of few hours to a few weeks (e.7 mmol/litre) occurred in the first 90 minutes of life. resuscitation of babies with positive pressure ventilation and low Apgar scores (less than 3) at 1 minute and 5 minutes. but serious. [EL = 3] Another cross-sectional study358 assessed types and frequencies of complications occurring in babies of women with diabetes. P = 0. Congenital malformations occurred in 7. Women with type 1 diabetes had higher rates of preterm birth (31% versus 10%. P = 0.002).354 Incidence of neonatal morbidities A cohort study355 assessed the effect of rigorous management of type 1 diabetes during pregnancy on perinatal outcome by comparing 78 pregnant women with type 1 diabetes managed prospectively with 78 matched controls who did not have diabetes.1 Neonatal care Initial assessment and criteria for admission to intensive or special care Description of the evidence A number of morbidities present in babies born to women with diabetes (including preexisting type 1 and type 2 diabetes and gestational diabetes).

The caesarean section rate was high. jitteriness. irritability. poor perfusion and/or absence of respirations in the baby should be assessed to determine whether endotracheal intubation and respiratory support is needed. If levels are 1.359[EL = 4] Hypoglycaemia: The prevention and treatment of neonatal hypoglycaemia is addressed separately in Section 7. the presence of hypotonia. Brachial plexus (Erb palsy).5% was high. hypoglycaemia. An enlarged heart or diffuse. the need for respiratory support and replacement surfactant therapy could be determined. Together with arterial blood gas results. hepatomegaly. Manifestations included fractures of the clavicle and/or humerus. seizures. The number of babies with asphyxia.359 As most fractures of this nature heal without treatment. In cases where a fracture is suspected. bilirubin levels should be monitored.2. a chest radiograph may be used to confirm the presence of fractures. 131 .354. The overall mortality rate of 7. an echocardiogram was prescribed in order to prevent congestive heart failure.354 [EL = 3] Severe asphyxia: If severe asphyxia has occurred at birth. seizures.75 mmol/litre or more.359 A review addressing medical concerns in the neonatal period described jaundice in the first 24 hours of life as pathological and requiring immediate evaluation and therapy as bilirubin in high concentrations is considered a cellular toxin. as the baby may show an asymmetric reflex or limited use of the arm on the affected side. The study recommended that women with diabetes should be offered regular antenatal care to maintain good glycaemic control during pregnancy. delayed feeding. Where left ventricular outflow obstruction was suspected. apnoea. the birth should be attended by experienced paediatricians to minimise complications and when there is clinical evidence of macrosomia caesarean section should be offered to reduce birth injuries. fine granular densities are consistent with respiratory distress syndrome. venous haematocrit measurements are indicated. These should indicate monitoring of serum calcium levels. calcium replacement should be started. decreased urine output). as was the rate of birth injuries among those who had vaginal birth. The startle reflex could also be used to confirm the presence of fractures. polycythaemia and birth trauma.Neonatal care admitted to a NICU and detailed maternal history and physical examination were performed to detect any congenital anomalies. The following evidence is drawn from two narrative non-systematic reviews. hypocalcaemia and hyperbilirubinaemia was high.359[EL = 4] Hypocalcaemia: Signs and symptoms of hypocalcaemia include coarse tremors. twitching. [EL = 2−] Neonatal assessment No clinical studies were identified that addressed the assessments that babies of women with diabetes should undergo. congenital anomalies. [EL = 4] Respiratory distress: Use of chest radiographs has been suggested for babies displaying signs of respiratory distress syndrome. phrenic nerve or cerebral injuries were also reported.359[EL = 4] Hyperbilirubinaemia: Where there has been lethargy. admission to a NICU is probably not necessary. irritability and seizures. hypoglycaemia.359 [EL = 4] Polycythaemia: For babies displaying clinical signs of polycythaemia (respiratory distress syndrome. feeding intolerance.359 [EL = 4] Fetal macrosomia: The investigation of birth trauma in macrosomic babies of women with gestational diabetes has been described as necessary.

03).0 mU/ml.361 Of the 4032 babies born in the study hospital.16 days compared with 3. 1023 were admitted to NICU. [EL = 2+] Neonatal hypoglycaemia: A prospective cohort study investigated the frequency and risk factors for neonatal hypoglycaemia and long-term outcomes of promptly treated neonatal hypoglycaemia. Low pO2 at birth was associated with amniotic fluid EPO (P < 0. The frequency of SGA babies was equal among the classes. Ninety-four (9. babies of women with gestational diabetes requiring no insulin had the lowest representation of LGA babies (25%). poor feeding or neonatal depression). admission to neonatal intensive care (P = 0.74 days for AGA babies). the highest rates (56%) were seen in women with type 1 diabetes which had been diagnosed before the age of 10 years.0001) and hyperbilirubinaemia (P = 0. 34. amniotic fluid EPO was the only variable to remain independently associated with low umbilical artery pH (P < 0.7 to 13.002) occurred significantly more often in the high EPO group than in the low EPO group. In 26. For the 182 babies (34%) presenting with respiratory distress syndrome of varying severity. EPO measurements were compared with those from 19 healthy. either within 2 days of birth or at birth. birthweight z-score. sepsis or the mother having diabetes were present.0001) and had significantly lower pO2 levels than those in the intermediate and low EPO groups (P < 0. P < 0. Amniotic fluid EPO levels above 63.0001).8% were SGA.0001). respiratory distress syndrome and higher birthweight: A retrospective cohort study conducted in the USA over a 3 year period acquired data for 530 babies born to 332 women with gestational diabetes and 177 women with type 1 diabetes. cardiac arrhythmias. lethargy. Newborn babies in the high EPO group were significantly more likely to be macrosomic (P = 0.1% of babies were born to women with pre-existing maternal diabetes or gestational diabetes and 12. maternal BMI.1% of the women with diabetes who were divided into three groups: low EPO less than 13.004).Diabetes in pregnancy Criteria for admission to intensive/special care Amniotic fluid erythropoietin: A case–control study investigated whether chronic fetal hypoxia. P < 0. Seventy-four (14%) of the 530 babies had macrosomia and 57% 132 . had had diabetes for 20 years or longer or had complications of diabetes. Of the 48 babies undergoing Bayley’s psychometric evaluations. non-smoking women delivered by elective caesarean section with an uneventful singleton pregnancy producing a healthy newborn baby. n = 155) than in control pregnancies (6.7.0001) and neonatal hypoglycaemia (P = 0.0 mmol/litre more than 6 hours after birth. ten were described as having miscellaneous conditions (apnoea.8–63.0 to 1975. [EL = 2+] Gestational age. or if risk factors such as SGA.1% of babies were preterm (37 weeks or less). gestational age at birth. while those born to women with type 1 diabetes which had been diagnosed before the age of 10 years and had a duration of 20 years had the highest rate (62%). Neonatal hypoglycaemia (blood glucose less than 2. apnoea. last amniotic fluid EPO level and last maternal HbA1c level. poor feeding or seizures were observed. LGA. SGA babies required the longest duration of intravenous dextrose infusion (5.8 mU/ml. The cohort was followed for 24 months. The median amniotic fluid EPO level was significantly higher in the women with diabetes (14. as indicated by amniotic fluid erythropoietin (EPO) levels. Similarly. during which time they were assessed neurologically and developmentally using the Bayley motor and developmental scales.002). cardiomyopathy (P < 0.360 An amniotic fluid sampled for EPO measurement was taken from 157 women with type 1 diabetes who gave birth by caesarean section before the onset of labour (one vaginal birth). The study found 47% (247) of babies were admitted to a NICU. range 2. After adjusting for the effects of maternal age.6% of the babies no known risk factors for hypoglycaemia were observed.0 mU/ml.3 mU/ml. was associated with neonatal complications in pregnancies complicated by type 1 diabetes. decreasing to 25% in babies of women with gestational diabetes requiring no insulin. range 1. 22 babies had congenital malformations. 103 babies with a gestational age of 34 weeks or more had respiratory distress syndrome and 32 babies had hypoglycaemia as the only diagnosis.0005) and acidotic (P < 0.0001).0001) and birthweight z-score (P = 0.0 mU/ml and high EPO more than 63.362 Seventy-six babies had a gestational age of 33 weeks or less. two showed a motor deficit at 6 months and one showed a language deficit at 24 months.0 mU/ml were considered to indicate fetal hypoxia and these elevated values were observed in 14.18%) were evaluated as having hypoglycaemia. The study found that 51. preterm birth. n = 19. Evaluations were performed if symptoms such as hypothermia. intermediate EPO 13.

Logistic regression analysis also showed that after controlling for gestational age and type of diabetes. The rate of emergency lower segment caesarean section in women with gestational diabetes or type 2 diabetes was 25%. 50 were AGA and two were SGA. Among the 74 babies who were macrosomic. The rate of emergency lower segment caesarean section of women with gestational diabetes or type 2 diabetes whose babies were admitted to NICU was 38% (52/136). three of polycythaemia. and the babies of women who had had diabetes for 20 years or longer or had complications of diabetes (38%). Neonatal hypoglycaemia was similar among babies born to women with gestational diabetes requiring no insulin (23%). [EL = 2+] Myocardial hypertrophy and respiratory distress syndrome: A cross-sectional study looked at the association between poorly controlled maternal diabetes and myocardial hypertrophy. Two of these babies died from cardiorespiratory failure within 48 hours of birth. or a duration less of than 10 years with no vascular lesions (25%). 25% (125) were treated for hyperbilirubinaemia and. Fifty-six percent (58/104) of the gestational diabetes was reclassified as normal. Ten babies were macrosomic and had myocardial hypertrophy as determined by echocardiograph. The most common indication for admittance to NICU was hypoglycaemia. 18% (43) were then transferred to the NICU (19 with respiratory distress syndrome as the main reason for transfer. The study found 46% (63/136) babies were delivered preterm (before 37 weeks).363 Twelve neonates were admitted to NICU with respiratory distress and cardiomegaly. 32 had hypoglycaemia. those with gestational diabetes requiring insulin (24%) and those with type 1 diabetes with age of onset 20 years or more. Among those delivered by caesarean section. IGT or type 2 diabetes after postpartum 75 g OGTTs. 61 were delivered at 33 weeks of gestation or less. of these. When a similar comparison was made for preterm birth the rates were 19% compared with 46% (63/136). Infant outcomes according to maternal antenatal and postpartum diagnoses were recorded. The presence of hyaline membranes at autopsy of the other two babies lends support to an association between respiratory distress syndrome and myocardial hypertrophy. seven for respiratory distress syndrome plus hypoglycaemia and one for poor feeding). Thirty of the 137 babies with hypoglycaemia were born before 34 weeks of gestation. The prevalence of neonatal hypoglycaemia was lower in these babies (P < 0. Women with gestational diabetes that was reclassified postpartum as IGT or type 2 diabetes accounted for the highest rates of preterm babies (86% [12/14] and 63% [12/19]. The study found 5% (13) of the 276 babies who had their haematocrit assessed were polycythaemic (haematocrit 0. Two babies were AGA and had cardiomegaly resulting from ventricular dilation in association with hypoglycaemia and acidaemia. respectively). 21 were also hypoglycaemic. Advanced maternal diabetes and lower gestational age were shown by logistic regression to be the strongest predictors of subsequent NICU care. there were 21 cases of hypoglycaemia. [EL = 3] Gestational age and mode of birth: A clinical audit was conducted at the National Women’s Hospital in New Zealand364 which serves a multi-ethnic population with a high background prevalence of type 2 diabetes. The rate of treatment for nondiabetic.Neonatal care (42/74) of this group were delivered by caesarean section. 10% had two or more episodes lasting several hours. Of the 530 babies.65 or more). In total 136 babies of women with diabetes were admitted to NICU. [EL = 2+] Blood glucose levels were recorded for 514 babies. full-term babies delivered during the same 2 year period was 5%. 80% (8/10) had clinical findings suggesting respiratory distress syndrome. 55 were LGA. Of the surviving babies. which was documented in 51% of the 133 . One or more hypoglycaemic episodes occurred in 27% (137) of these babies.05) than the babies of women with type 1 diabetes with age of onset 10–19 years or duration 10–19 years with no vascular lesions (35%). one of hypocalcaemia and 12 of hyperbilirubinaemia. Thirty percent of macrosomic babies had respiratory distress syndrome. the babies of women with type 1 diabetes which had been diagnosed before the age of 10 years. Of the 244 babies admitted to well baby nurseries for routine care. Tweny-nine percent (112/382) of the babies of women with gestational diabetes were admitted and 40% (24/60) of the babies of women with type 2 diabetes were admitted. While 90% of the babies responded rapidly to treatment. Of the 244 babies (46% of total group) assigned to ‘well baby nurseries’ for routine care and enteral feeding. breastfed babies were more likely to succeed with routine care and enteral feeding. 16 for treatment of hypoglycaemia.

32 [EL = 3] Existing guidance The NSF for diabetes20 recommends that ‘Neonatal intensive care is only indicated for babies who display persistent hypoglycaemia after 3 hours of age.34).Diabetes in pregnancy babies. hypoglycaemia. but two narrative non-systematic reviews described the clinical signs of the most frequently occurring neonatal complications in babies of women with diabetes. One of the observational studies reported that persistent or recurrent hypoglycaemia in the neonatal stage can lead to neurodevelopmental deficits later in life. although babies of women with diabetes are believed to be at increased risk of these complications. This was followed by respiratory distress in 40% of babies. hypoglycaemia.2) and recommended that all units where women with diabetes give birth should have written policy for management of the baby and that the policy should assume that babies will remain with their mothers in the absence of complications. birth trauma (e. or before the first three or four feedings. asymptomatic hypoglycaemia in a healthy baby 26% (11/42) and a clinical need for admission such as poor feeding or respiratory problems 43% (18/42). A third of women with type 2 diabetes antenatally or postpartum had babies weighing more than 4000 g. Rates of respiratory distress in the preterm babies and term babies were not significantly different (39% [26/67] versus 43% [31/70]. The care plan was not fully followed for 35% (18/51) of babies remaining with their mothers. None of the studies reported incidence of hypoxic ischaemic encephalopathy or hypomagnesaemia.05) and significantly more common than in the IGT or normal group (P < 0. Complications reported in these studies included asphyxia. Wales and Northern Ireland reported that 30% (64/213) had a policy of routinely admitting babies of women with diabetes to the neonatal or special care unit. shoulder dystocia). macrosomia. [EL = 2+] Current practice The CEMACH enquiry covered neonatal care of term babies born to women with pre-existing type 1 or type 2 diabetes. P = 0.05). 12/24 babies). transitional care unit. hyperbilirubinaemia. aspects of the care plan that were not followed included blood glucose management. feeding and temperature. A further four observational studies and a clinical audit investigated neonatal complications (including fetal hypoxia. especially in the area of feeding (50%. and in the presence of clinical signs of hypoglycaemia. 134 . hypocalcaemia.g. respiratory distress syndrome and associated mortality. The three main indications for admission to a NICU were a hospital policy of routine admission of healthy babies of women with diabetes 29% (12/42). The authors of the study recommended that high-risk babies be screened at regular intervals in the first 48 hours of life if not being fed. stayed on the labour ward or in a maternal dependency unit and 42 were admitted to a NICU for special care. polycythaemia and respiratory distress syndrome) and indications for admission to a NICU for babies of women with pre-existing type 1 or type 2 diabetes and gestational diabetes. The enquiry panels assessed that 57% (24/42) of the admissions were unavoidable and that subsequent care of 63% (15) of the babies was compromised. These birthweights were significantly higher than for the IGT group (P < 0. No clinical studies were identified in relation to neonatal assessment that babies of women with diabetes should undergo. myocardial hypertrophy (hypertrophic cardiomyopathy). hypocalcaemia. LGA. 70 were admitted to a postnatal ward.33 In the 112 babies selected for the neonatal enquiry that had medical records available. The enquiry also discussed the importance of early skin-to-skin contact between babies and their mothers (see Section 7. [EL = 3–4] A 2002 CEMACH audit of units expected to provide maternity care for women with diabetes in England.’ Evidence statement Five observational studies have reported on the incidence of neonatal morbidity in babies born to women with diabetes. There was evidence of a clear written care plan for 73% (51/70) of babies who remained with their mothers and 57% (24/42) of babies admitted to a NICU. congenital malformations.

depending on local guidelines. hypomagnesaemia. blood tests for polycythaemia. The NSF for diabetes20 recommends that admission to a NICU should be made only for babies with persistent hypoglycaemia. congenital malformations (cardiac and central nervous system). and respiratory distress (several of which are potentially life-threatening) are more prevalent in babies of women with pre-existing diabetes and gestational diabetes. therefore. Some babies with clinical signs of the conditions listed above may be cared for in a transitional care unit. hypoglycaemia. Babies of women with diabetes should be kept with their mothers unless there is a clinical complication or abnormal clinical signs that warrant admission for intensive or special care. the CEMACH audit reported that 30% of units still routinely admit babies of mothers with diabetes to the neonatal or special care unit and that the most frequent reasons for admission to a NICU were routine policy and asymptomatic hypoglycaemia. 135 . However. myocardial hypertrophy (hypertrophic cardiomyopathy). Early community midwifery support for these babies should be more intense than average. The clinical audit reported that the most frequent indications for admission to NICU were hypoglycaemia and respiratory distress syndrome. polycythaemia and hyperviscosity. thus avoiding unnecessary investigations. which will represent cost savings to the NHS and should provide reassurance for parents.2). to keep babies with their mothers except when there is a clinical reason to separate them. Neonatal metabolic adaptation in babies of women with diabetes is generally completed by 72 hours of age. and one of the observational studies reported that the prevalence of these complications was higher with increasing duration of diabetes. babies with these conditions should be admitted to a neonatal unit. be competent to recognise and manage these conditions and women with diabetes (including gestational diabetes) should be advised to give birth in hospitals where advanced neonatal resuscitation skills are available 24 hours a day. neonatal encephalopathy. represents a cost saving to the NHS.Neonatal care Other observational studies showed that prematurity and birth by emergency caesarean section were predictors for NICU admission in women with type 2 diabetes and those with gestational diabetes. hypocalcaemia. From evidence to recommendations Evidence shows that birth trauma.20 thus bringing cost savings to the NHS and maximising the opportunity for early skin-to-skin contact between babies and their mothers and initiation of breastfeeding (see Section 7. hypocalcaemia and hypomagnesaemia. The GDG’s view is that blood glucose testing should be carried out routinely (at 2–4 hours after birth) for babies of women with diabetes because of the risk of complications arising from asymptomatic hypoglycaemia (see Section 7. Cost-effectiveness The effectiveness of criteria for admission to neonatal intensive/special care for babies of women with diabetes was identified by the GDG as a priority for health economic analysis. facilities and care pathways. Healthcare professionals assessing such babies should. and investigations for congenital heart malformations and cardiomyopathy should be reserved for babies with clinical signs of these complications. hyperbilirubinaemia. Thus no health economic modelling is needed to demonstrate that reinforcing the NSF recommendation. in accordance with the recommendations contained in the NSF for diabetes.2). Several of the studies suggested that babies of women with diabetes should be closely monitored and admitted to intensive care only in unavoidable circumstances where there are clinical signs of hypoglycaemia and/or respiratory distress. Transfer to community care is not recommended before 24 hours and not before healthcare professionals are satisfied that the baby is maintaining blood glucose levels and has developed good feeding skills because of the risk of recurrent hypoglycaemia in the early neonatal period. However. Where such facilities are unavailable. hyperbilirubinaemia. thus avoiding unnecessary separation of mothers and babies.

Blood tests for polycythaemia. and not before healthcare professionals are satisfied that the babies are maintaining blood glucose levels and are feeding well. hypocalcaemia and hypomagnesaemia should be carried out for babies with clinical signs. Close surveillance should be maintained in babies with risk factors for compromised metabolic adaptation if the plasma glucose concentration is less than 2.365. babies with abnormal clinical signs. There were no research recommendations relating to the initial assessment of babies and criteria for admission to intensive/special care. Babies of women with diabetes should not be transferred to community care until they are at least 24 hours old. Blood glucose testing should be carried out routinely in babies of women with diabetes at 2–4 hours after birth.0 mmol/litre.4 mmol/litre) an intravenous glucose infusion is indicated to raise the glucose level above 2. 7. [EL = 2+] A consensus statement368 discussed the definition of neonatal hypoglycaemia. Babies of women with diabetes should be kept with their mothers unless there is a clinical complication or there are abnormal clinical signs that warrant admission for intensive or special care.193. but rather to indicate the level at which intervention (additional feeding and.1–1. It is based on a study that found adverse neurodevelopmental outcomes to be associated with repeated values below this level.6 mmol/litre. babies with risk factors for compromised metabolic adaptation. preterm babies and babies receiving parenteral nutrition. Babies of women with diabetes should have an echocardiogram performed if they show clinical signs associated with congenital heart disease or cardiomyopathy. at very low concentrations (1.367 The study involved 661 preterm babies and used multiple regression to show that reduced developmental scores were associated independently with plasma glucose concentration less than 2.366 This threshold is not used to diagnose the condition. generally asymptomatic.6 mmol/ litre. Babies of women with diabetes should be admitted to the neonatal unit if they have: • • • • • • • • • hypoglycaemia associated with abnormal clinical signs respiratory distress signs of cardiac decompensation due to congenital heart disease or cardiomyopathy signs of neonatal encephalopathy signs of polycythaemia and are likely to need partial exchange transfusion need for intravenous fluids need for tube feeding (unless adequate support is available on the postnatal ward) jaundice requiring intense phototherapy and frequent monitoring of bilirubinaemia been born before 34 weeks (or between 34 and 36 weeks if dictated clinically by the initial assessment of the baby and feeding on the labour ward).2 Prevention and assessment of neonatal hypoglycaemia Description of the evidence The working definition for neonatal hypoglycaemia is blood glucose less than 2.Diabetes in pregnancy Recommendations for initial assessment and criteria for admission to intensive or special care Women with diabetes should be advised to give birth in hospitals where advanced neonatal resuscitation skills are available 24 hours a day. intravenous dextrose) should be considered. non-recurrent and good response to intravenous dextrose. The timing of the examination will depend on the clinical circumstances. including heart murmur. The statement considered term babies. [EL = 4] The characteristics of neonatal hypoglycaemia in babies of women with diabetes are very early onset (first hour after birth).368 136 .5 mmol/litre. if this does not reverse the hypoglycaemia. hyperbilirubinaemia.

Neonatal care

Early feeding Two studies were found that investigated the effect of timing of first feed on blood glucose levels. The studies were undertaken in the 1960s when delaying the initial feed was common.
The first study compared 27 preterm babies allocated to an ‘early fed’ group (fed with formula from 6 hours of age) with 41 babies fasted for 72 hours.369 At 72 hours 24/41 babies in the fasted group had blood glucose levels below 1.4 mmol/litre. In the early fed group no babies had blood glucose values below this level. Statistical significance was not reported. [EL = 2+] The second study compared 118 preterm babies fed at 3 hours with undiluted breast milk with 121 fed at a later stage, usually at 12 hours. There were no cases of symptomatic hypoglycaemia in the early fed group compared with four cases in the later fed group. Blood sugar estimation was introduced in phase three of the trial. The lowest level was less than 1.1 mmol/litre in 5/44 in the ‘immediate fed’ group compared with 10/54 in the ‘later fed’ group.370 Statistical significance was not reported. [EL = 2+]

Frequent feeding One study was identified that looked at the effect of frequency of initial feeds on blood glucose levels.
The study was a cross-sectional study of 156 term babies.371 A multiple regression analysis with method of feed, between-feed interval, volume of feed and postnatal age as independent variables found only between-feed interval (minutes) to be significantly correlated with blood glucose concentration (B = −0.003, SE = 0.001, β = −0.32, P < 0.05). [EL = 2+]

Breastfeeding Ten studies were found that had implications for choice of feeding method.
The first study compared 45 breastfed babies with 34 formula-fed babies.372 The babies were 6 days old and matched for gestation and birthweight. Breastfed babies had significantly higher levels of ketones. [EL = 2+] The second study compared 71 breastfed babies with 61 formula-fed babies.371 All babies were term babies less than 1 week old. Breastfed babies had significantly lower mean blood glucose concentration (P < 0.05) and significantly higher ketone body concentrations (P < 0.001). Breastfed babies had higher total gluconeogenic substrate concentrations (P < 0.01). [EL = 2+] A cohort study investigated the glucose concentration of breast milk of women with diabetes and its relationship with the quality of metabolic control.373 The study involved 11 women with type 1 diabetes and 11 age-matched women without diabetes. The women with diabetes had intensified insulin treatment and their average HbA1c values were significantly higher than those in women without diabetes (8.1 ± 0.9% versus 6.2 ± 0.5%, P < 0.01). The glucose concentration of breast milk taken from women with diabetes did not differ from that of women without diabetes (0.68 ± 0.50 versus 0.66 ± 0.55 mmol/litre). No correlation was found between the maternal blood glucose (HbA1c) and the glucose concentration of breast milk. [EL = 2−] A prospective cohort study374 investigated whether children born to women with diabetes were at increased risk of developing obesity and IGT in childhood. A total of 112 children of women with diabetes (type 1 diabetes, n = 83 and gestational diabetes, n = 29) were evaluated prospectively for impact of ingestion of either breast milk from a woman with diabetes or banked donor breast milk from women without diabetes during the early neonatal period (days 1–7 of life) on relative body weight and glucose tolerance at a mean age of 2 years. There was a positive correlation between the volume of breast milk from women with diabetes ingested and risk of overweight at 2 years of age (OR 2.47, 95% CI 1.25 to 4.87). In contrast, the volume of banked donor breast milk from women without diabetes ingested was inversely correlated to body weight at follow-up (P = 0.001). Risk of childhood IGT decreased by increasing amounts of banked donor breast milk ingested neonatally (OR 0.19, 95% CI 0.05 to 0.70). Stepwise regression analysis showed volume of breast milk from women with diabetes to be the only significant predictor of relative body weight at 2 years of age (P = 0.001). The results suggest that early neonatal ingestion of breast milk from women with diabetes may increase the risk of becoming overweight and, consequently, developing IGT during childhood. [EL = 2+]

137

Diabetes in pregnancy

A prospective cohort study375 investigated whether intake of breast milk of women with diabetes during the late neonatal period and early infancy influenced subsequent risk of overweight (adipogenic) and IGT (diabetogenic) in children born to women with diabetes. One hundred and twelve children born to women with diabetes were evaluated for influence of ingesting their mother’s breast milk during the late neonatal period (second to fourth neonatal week) and early infancy on relative body weight and glucose tolerance in early childhood. Exclusive breastfeeding was associated with increased childhood relative body weight (P = 0.011). Breastfed children of women with diabetes had an increased risk of overweight (OR 1.98, 95% CI 1.12 to 3.50). Breastfeeding duration was also positively related to childhood relative body weight (P = 0.004) and 120 minute blood glucose during an OGTT (P = 0.022). However, adjustment for the volume of breast milk from women with diabetes ingested during the early neonatal period (i.e. the first week of life), eliminated all these relationships with late neonatal breastfeeding and its duration. No relationship was observed between maternal blood glucose in the middle of the third trimester and neonatal outcomes. The study suggests that neither late neonatal breast milk intake from women with diabetes nor duration of breastfeeding has an independent influence on childhood risk of overweight or IGT in children born to women with diabetes. The first week of life appears to be the critical window for nutritional programming in children of ingestion of breast milk from women with diabetes. [EL = 2+] Another cohort study376 investigated whether late neonatal ingestion of breast milk might independently influence neurodevelopment in 242 children of women with diabetes. There was no impact of ingestion of breast milk of women with diabetes on psychomotor parameters, but it negatively influenced onset of speaking with children of women with diabetes who were fed solely on breast milk taking the longest time to initiate speech. Adjusting for the amount of breast milk ingested during the early neonatal period weakened the hazard ratio towards nonsignificance. The data suggest that neonatal ingestion of breast milk of women with diabetes, particularly during the first week of life, may delay speech development, an important indicator of cognitive development. [EL = 2++] Another cohort study investigated the extent to which early breastfeeding or exposure to cow’s milk affected psychomotor and cognitive development in children of women with diabetes.377 Children of women with diabetes with early breast milk ingestion achieved early psychomotor developmental milestones (lifting head while prone, following with eyes; P = 0.002). However, children who had ingested larger volumes of milk of women with diabetes had a delayed onset in speaking compared to those with lower milk intake (P = 0.002). The data suggest that ingesting larger volumes of milk of women with diabetes may normalise early psychomotor development in babies of these women, but may delay onset of speaking. [EL = 2++] A systematic review378 summarised the clinical evidence relating a short duration of breastfeeding or early cow’s milk exposure to the development of type 1 diabetes. People with type 1 diabetes were more likely to have been breastfed for less than 3 months during their infancy (pooled OR 1.43, 95% CI 1.15 to 1.77) and to have been exposed to cow’s milk before 4 months (pooled OR 1.63, 95% CI 1.22 to 2.17) compared to those without diabetes. The study suggests that early exposure to cow’s milk may be an important determinant of subsequent type 1 diabetes and may increase the risk approximately 1.5 times. [EL = 2++] Another systematic review379 evaluated the relationship between early infant diet and the risk of developing type 1 diabetes in later life via a meta-analysis of 17 case–control studies involving 21 039 people who were either breastfed or introduced early to cow’s milk. The effect of exposure to breast milk substitutes on developing type 1 diabetes was small. [EL = 2++] A case–control study investigated the association between the type of feeding in infancy and the development of type 1 diabetes.380 The study involved 100 children with type 1 diabetes and 100 children without diabetes matched for sex and age. Information on feeding patterns during the first year of life was collected using a questionnaire. A larger proportion of children with diabetes had been breastfed. There was no clear difference between children with diabetes and those without diabetes in terms of duration of breastfeeding (children with diabetes, median duration 3 months; children without diabetes, median duration 2 months). The data do not support the existence of a protective effect of breastfeeding on the risk of type 1 diabetes, or that early exposure to cow’s milk and dairy products influences the development of type 1 diabetes. [EL = 2+]

138

Neonatal care

Barriers to breastfeeding in women with diabetes in pregnancy One study was identified that compared breastfeeding initiation and maintenance in 33 women with type 1 diabetes to those of 33 women in a control group and 11 women in a reference sample.381 The control group consisted of women without diabetes selected using gestational age at delivery, method of delivery, sex of baby and prior lactation experience. The reference group consisted of women without diabetes who were within 90–110% of ideal body weight prior to conception, had uncomplicated pregnancies and delivered vaginally. The study found women with diabetes were more likely to experience difficulties establishing and continuing breastfeeding than control and reference groups. All differences were significant (P < 0.05). The difference between groups was attributed to differences in postpartum care. Hospital protocol placed all babies of women with diabetes in the neonatal unit after birth for monitoring for hypoglycaemia. This meant that women with diabetes saw their babies the least amount of time in the first 3 days postpartum, waited the longest to begin breastfeeding their babies and breastfed their babies fewer times. Other possible contributory factors were that 70% of the women with diabetes had undergone caesarean section and that 30% of the babies of women with diabetes were macrosomic. Women with diabetes cited baby sleepiness as the most common babyfeeding problem. A sleepy baby was not identified as a problem by any of the women in the control group and by only one woman in the reference group. [EL = 2+]
A case–control study382 investigated factors influencing the initiation and maintenance of breastfeeding in 22 women with type 1 diabetes and 22 women without diabetes. Diabetes was not a principal factor in the decision to breastfeed or bottle-feed for the majority of the women. Women who considered diabetes in their decision to breastfeed had on average 2 years more of education than those who did not (14.82 years versus 12.94 years). Although the women did not perceive diabetes as influencing their breastfeeding experiences, they found that maintaining good control of diabetes required greater effort and flexibility during breastfeeding. [EL = 2+]

Banking colostrum before birth Two publications were identified in relation to production of colostrum from women with diabetes and banking colostrum before birth for use in the neonatal period.
A cohort study compared the composition of macro- and micronutrients in milk from six women with tightly controlled type 1 diabetes (median glycosylated haemoglobin concentrations at parturition of 5.2% (range 4.9–5.3%) and 6 weeks later of 6.1% (range 5.0–6.3%), reference range 5.0–6.4%) with that from five women without diabetes.383 Milk samples were collected halfway through a single breastfeed at days: 3–5 (colostrum); 7, 9 and 10 (transitional milk); and 12, 15, 17, 21, 25, 29 and 35 (mature milk). There were no differences between the two groups in terms of concentrations of macronutrients (triglycerides, lactose and protein), cholesterol, glucose or myoinositol, nor in fatty acid composition. The duration of colostrum lactation was the same for women with diabetes and those without diabetes (3–5 days in both groups). [EL = 2−] A narrative non-systematic review considered expressing and banking colostrum antenatally for use in the neonatal period.384 The review suggested that women with conditions that may delay breastfeeding and those who wish to lessen known familial health problems for their expected babies (including women with type 1 diabetes or gestational diabetes) would benefit from antenatal expression of colostrum. The risk of nipple stimulation initiating oxytocin release and, therefore, preterm contractions, labour and preterm birth was discussed and a protocol for expressing and storing colostrum was suggested. The review concluded that expressing and storing colostrum is advantageous to babies and confidence building for women and should, therefore, be supported for any condition which healthcare professionals consider to be relevant. [EL = 4]

Testing for neonatal hypoglycaemia A systematic review by the WHO366 found that screening for hypoglycaemia using glucose oxidase-based reagent strips had poor sensitivity and specificity. The report recommended that ‘less frequent but more accurate laboratory or ward-based glucose electrode measurements among babies at risk are preferable’.

139

Diabetes in pregnancy

Intravenous dextrose for neonatal hypoglycaemia There is a consensus that intravenous dextrose should be administered for symptomatic hypoglycaemia and for asymptomatic hypoglycaemia that fails to respond to feeding.366,386,387 However, no clinical studies were identified in relation to evaluation of protocols for the treatment of neonatal hypoglycaemia using intravenous dextrose. Current practice The CEMACH enquiry33 reported that the opportunity for early skin-to-skin contact after birth was achieved in 29% (30) of the 102 babies whose medical records were available. In eight cases, skin-to-skin contact was not possible due to the condition of the woman and/or the baby. Ninety-five percent of babies remaining with their mothers received their first feed on the labour ward compared with 50% of those admitted to a neonatal unit (P < 0.001). Twenty-six percent (29/112) of women received help with breastfeeding within 1 hour of birth (34% of women on labour wards and 12% of women in the neonatal unit). Thirty-one percent of women whose babies were admitted to the neonatal unit had documented evidence in their medical records that they were shown how to breastfeed and maintain lactation. Infant formula was given at the first feed for 63% (67/106 babies) and this was the first choice for women in 46% (32/70) of cases. Breast milk was the first feed for 50% (34/68) of babies that remained with their mothers and 21% (8/38%) of babies in the neonatal unit (P = 0.001). The first feed given was not the mother’s intended type of feed for 28% (27/96) of babies (16% of women who stayed with their babies and 50% of those admitted to the neonatal unit, P < 0.001). [EL = 3–4]
CEMACH undertook a descriptive study of all pregnancies of women with pre-existing diabetes who gave birth or booked between 1 March 2002 and 28 February 2003.2 The study found that 40.1% of all babies (1382/3451) were fed within 1 hour and 78.8% (2717/3451) by 4 hours. Among term babies 46.5% (1031/2216) were fed within 1 hour and 87.7% (1837/2216) within 4 hours. Exclusive breastfeeding was the choice at birth for 53% (1762/3342) of women with pre-existing diabetes compared with 69% in the general population. At 28 days after birth the proportion of exclusively breastfed babies was 23.8%, half the proportion who had intended to breastfeed at birth. A history of low blood glucose alone was the main reason (36.7%) for giving term babies of women with diabetes supplementary milk or glucose. In 9% of cases babies were given supplementary milk or glucose routinely according to local practice, possibly compromising establishment of breastfeeding. Of the 3451 babies in the study, 83.2% were tested within 6 hours and 47.3% were tested within 1 hour. Testing this early may, however, simply detect the normal drop in blood glucose that can be expected after birth. One-third of term babies were admitted to a neonatal unit for special care. Examining the reasons for admission suggested that many (67%) were avoidable. [EL = 3] The CEMACH enquiry33 reported that neonatal blood glucose testing was mainly carried out using reagent strips. It supported the WHO’s recommendation that reagent strip testing is unreliable and recommended that when considering the diagnosis of hypoglycaemia at least one laboratory value should be obtained. The enquiry also recommended that women with diabetes should be informed antenatally of the beneficial effects of breastfeeding on metabolic control for them and their babies and that blood glucose testing performed too early should be avoided in well babies without signs of hypoglycaemia. [EL = 3–4] A standard textbook of neonatology388 supports this evidence.

Existing guidance
The NSF for diabetes advises that babies born to women with diabetes should be fed as soon as possible after birth.20 It also recommends breastfeeding for babies of women with diabetes, but that women should be supported in the feeding method of their choice. [EL = 4] The NICE guideline for routine postnatal care recommends that women should be encouraged to have skin-to-skin contact with their babies as soon as possible after birth and that initiation of breastfeeding should be encouraged as soon as possible after birth and ideally within 1 hour.11

140

Neonatal care

Evidence statement
The blood glucose concentration used to guide intervention for neonatal hypoglycaemia (i.e. additional feeding and, if this does not reverse hypoglycaemia, intravenous administration of dextrose) is 2.6 mmol/litre. Close surveillance should be maintained in babies with risk factors for compromised metabolic adaptation if the plasma glucose concentration is less than 2.0 mmol/litre; at very low concentrations (1.1–1.4 mmol/litre) an intravenous glucose infusion is indicated to raise the glucose level above 2.5 mmol/litre. Two studies showed that early feeding of babies was associated with lower incidence of hypoglycaemia than late feeding (more than 12 hours after birth). However, these studies involved preterm babies who may demonstrate different metabolic adaptation to term babies. Another study showed between-feeding interval to be correlated with blood glucose levels, suggesting that frequent feeding should be encouraged to prevent neonatal hypoglycaemia. Three studies relating to choice of infant-feeding method for women with diabetes suggested that: breastfeeding may enhance ketogenesis and that ketones may be an important alternative to glucose for brain metabolism in the neonatal period; breastfeeding babies of women with diabetes was not associated with increased exposure of the babies to high glucose levels; and, where possible, separation of mother and baby should be avoided to enable early feeds, frequent feeds and breastfeeding, with the possibility that supplementary feeding with infant formula may be required for women with diabetes who breastfeed. However, the first of these studies involved 6-day-old babies and therefore has limited relevance to hypoglycaemia in babies of women with diabetes, who are at greatest risk of hypoglycaemia in the first 12 hours. A further seven observational studies, including two systematic reviews of observational studies, examined associations between feeding method and long-term outcomes. Three of the studies showed that obesity, IGT and impaired cognitive development were associated with ingestion of breast milk from women with diabetes. However, the two systematic reviews, which showed an association between breastfeeding and subsequent development of diabetes, were not specific to children of women with diabetes. Two studies reported that initiation and maintenance of breastfeeding was more difficult for women with diabetes because of routine separation of babies from their mothers at birth or clinical reasons for separation such as the woman having undergone caesarean section or the baby having macrosomia. Although diabetes was not a major factor in deciding whether to breastfeed, women with diabetes found that maintaining good control of diabetes required greater effort and flexibility during breastfeeding. These findings suggest that, where possible, separation of the mother and baby should be avoided to facilitate early, frequent feeds and breastfeeding. Supplementary feeding with infant formula may be required for women with diabetes who breastfeed. No clinical studies were identified in relation to the potential benefits of expressing and storing colostrum antenatally for the purposes of supporting early feeding to prevent hypoglycaemia in babies of women with diabetes. A systematic review by the WHO noted low sensitivity and specificity of reagent strip blood glucose testing to identify neonatal hypoglycaemia and recommended laboratory or ward-based glucose electrode measurements for babies at risk of neonatal hypoglycaemia. No clinical studies were identified in relation to the evaluation of protocols for treatment of neonatal hypoglycaemia using intravenous dextrose.

From evidence to recommendations
In the absence of high-quality evidence, the GDG’s recommendations for the prevention and treatment of neonatal hypoglycaemia are based on group consensus. The GDG’s view is that all maternity units should have a local written protocol for the prevention, detection and management of hypoglycaemia in babies of women with diabetes. Breastfeeding is recommended to prevent neonatal hypoglycaemia (by promoting successful metabolic adaptation) alongside other known benefits. Early commencement of breastfeeding is more important in babies of women with diabetes because of the risk of neonatal hypoglycaemia and is encouraged by skin-to-skin 141

Randomised controlled trials are needed to determine whether this practice is clinically and cost-effective. These factors should be explored in the randomised controlled trials. Babies of women with diabetes who present with clinical signs of hypoglycaemia should have their blood glucose tested and be treated with intravenous dextrose as soon as possible. Babies of women with diabetes should not be treated with invasive procedures (such as tube feeding or intravenous dextrose) unless they have clinical signs of hypoglycaemia or unless their blood glucose values persist below the threshold for initiating intravenous dextrose.0 mmol/ litre on two consecutive readings. There have been no clinical studies to evaluate the effectiveness of antenatal banking of colostrum in women with diabetes.6 mmol/litre. The GDG’s view is that blood glucose should be tested before feeding the baby. There is also a putative risk of precipitating uterine contractions through antenatal expression of colostrum and an accompanying release of oxytocin. be of benefit to babies of women with diabetes.Diabetes in pregnancy contact. Recommendations for prevention and assessment of neonatal hypoglycaemia All maternity units should have a written policy for the prevention. Babies of women with diabetes should have their blood glucose tested using a quality-assured method validated for neonatal use (ward-based glucose electrode or laboratory analysis). if there are abnormal clinical signs or if the baby will not feed orally effectively. If blood glucose values are below 2. 142 . feed as soon as possible after birth and at frequent intervals thereafter. While the target level for blood glucose is 2. Babies of women with diabetes should. Blood glucose measurements should be obtained using ward-based glucose electrode or laboratory analysis because these have greater sensitivity and specificity than reagent strip testing. therefore. Additionally. therefore.0 mmol/litre on two consecutive readings despite maximal support for feeding. In making this recommendation the GDG noted the findings of the CEMACH enquiry. detection and management of hypoglycaemia in babies of women with diabetes. Additional measures should only be implemented if one or more of these criteria are met. Babies of women with diabetes should feed as soon as possible after birth (within 30 minutes) and then at frequent intervals (every 2–3 hours) until feeding maintains pre-feed blood glucose levels at a minimum of 2. Encouraging women with diabetes to express and store colostrum before birth might be viewed as an additional barrier to breastfeeding in this group of women who already have lower breastfeeding rates than the general maternity population. Antenatal expression and storage of colostrum may. which reported that reagent strip testing is still commonplace. despite maximal support for feeding. additional measures such as tube feeding or intravenous dextrose should be given.0 mmol/litre. the opportunity for early skin-to-skin contact and initiation of breastfeeding is not always achieved in pregnancies complicated by diabetes because of the increased risk of neonatal complications requiring admission to intensive/special care. the GDG has set the threshold for initiating intravenous administration of dextrose at 2. Research recommendations for prevention and assessment of neonatal hypoglycaemia Is systematic banking of colostrum antenatally of any benefit in pregnancies complicated by diabetes? Why this is important Babies of women with diabetes are at increased risk of neonatal hypoglycaemia and may need frequent early feeding to establish and maintain normoglycaemia.

391 Of these. with the exception of mean blood glucose values during the first week postpartum which reached borderline significance (6. Fewer hypoglycaemic episodes in breastfeeding women were associated with breastfeeding sessions (4. In all women mean blood glucose values were significantly lower during the first week postpartum (6. Two cohort studies were identified that considered the effects of oral hypoglycaemic agents on breast milk and infant hypoglycaemia. nine women stopped breastfeeding before 6 weeks and 14 women bottle-fed. Women were advised to keep to regular meal times and eat before the baby’s feeding times.2.1 to 9.389 Breastfeeding was initiated by 15 women in the first 24 hours.56 ± 0. Five women with type 2 diabetes and two women without diabetes were started on 143 . Insulin dosages did not differ between the three groups. P < 0. At 7 days and 1 month postpartum 28 women were breastfeeding and at 2 months postpartum 24 women were breastfeeding. [EL = 3] Oral hypoglycaemic agents A cohort study392 investigated excretion of metformin into breast milk and the effect on nursing babies.5) than at other times (12. There was no difference in glycaemic control or insulin requirements between breastfeeding and bottle-feeding women.001). respectively).1 mmol/litre) than at preconception (7. the first 7 days postpartum.7 ± 1.050). In all women insulin requirements were significantly lower during the first week postpartum (0. P = 0.8 8.1 ± 2.11 U/kg/day.6 ± 1. [EL = 2+] The second cohort study followed 30 women with type 1 diabetes from birth to 6 weeks postpartum. If this was not possible.7 ± 0. the first month postpartum and the second month postpartum. On discharge from hospital women were prescribed an insulin regimen two-thirds of the third-trimester requirements and advised to adjust pre-meal insulin doses to glycaemic response.6 units (26%) from their pre-pregnancy dose (95% CI 8.1. Six week postpartum FBG levels were significantly lower in women who breastfed exclusively (4.16 U/kg/day) and they remained significantly lower over the first and second months postpartum (0.0 ± 0.9 mmol/ litre.56 ± 0.3 mmol/litre). Women who bottle-fed (n = 6) reduced their insulin dose by a mean of 5. The percentage of blood glucose readings below 3 mmol/litre did not differ during the four periods.2 ± 7.3 units. Insulin doses were reduced below the pre-pregnancy dose immediately after birth and then adjusted according to blood glucose concentrations.6 ± 2.0 ± 3. [EL = 2+] The case series involved 24 women with type 1 diabetes.9 mmol/ litre) or during the second month postpartum (7.68 ± 0.3%) from their prepregnancy dose (95% CI 1.6 ± 0. The study compared glycaemic control and insulin requirements of breastfeeding and bottle-feeding women over four periods: preconception.2 mmol/litre) compared with those in women who stopped breastfeeding before 6 weeks (8. not significant).9 to 14.7 mmol/litre). P = 0. 18 established breastfeeding and 16 continued until the 6 week postnatal clinic.15 U/kg/day) than at preconception (0. Factors affecting the choice between breastfeeding and bottle-feeding for babies of women with diabetes are considered in Section 7. Insulin The first cohort study involved 36 women with type 1 diabetes.6 versus 7. women were advised to have a glass of juice (during the day) or milk (at night).7 ± 1.002).3 units.56 ± 0.1 mmol/litre) and women who bottle-fed (6.390 Six women breastfed exclusively. After birth women who breastfed (n = 18) reduced their insulin dose by a mean of 11.15 U/kg/day and 0.2 units (11.1 Postnatal care Breastfeeding and effects on glycaemic control Description of the evidence Two small cohort studies and a case series were identified that considered the effect of breastfeeding on glycaemic control in women with diabetes.

The manufacturers state that enalapril is probably present in breast milk in an amount too small to be harmful. perindopril and ramipril have no information available and so the manufacturer advises women who are breastfeeding to avoid them. Four women dropped out. The average metformin concentration was under 0. Angiotensin-converting enzyme inhibitors A reference guide to drugs in pregnancy and lactation reported that there are limited data for the use of the ACE inhibitors enalapril and trandolapril.77 Although no studies have investigated their use in women who are breastfeeding. cilazapril. rosiglitazone. pioglitazone and rosiglitazone-related material may be present in low levels in breast milk.3% of the maternal weight-adjusted dose. Cilazapril. Eight women who received a single oral dose of 5 mg or 10 mg glibenclamide were studied by measuring drug concentrations in maternal blood and breast milk for 8 hours after the dosing schedule. The results suggest that glibenclamide and glipizide are safe and compatible with breastfeeding at the doses investigated. nateglinide. moexipril hydrochloride.77 [EL = 3] There was no information about gliclazide or gliquidone in the reference guide. the reference guide suggested nateglinide. chlorpropamide and tolbutamide are potentially toxic to babies if they are taken by breastfeeding women. The British National Formulary reports that metformin is present in breast milk and the manufacturer advises women who are breastfeeding to avoid it. captopril. The reference guide suggested that glimepiride and glipizide are likely to be present in breast milk. but the reference guide suggested that they are probably compatible with breastfeeding.78 The manufacturers of nateglinide. glibenclamide. and that data on safety during breastfeeding are needed. fosinopril sodium. perindopril or quinapril in women who are breastfeeding. [EL = 2+] A reference guide to drugs in pregnancy and lactation reports that women taking metformin can breastfeed. glimepiride and glipizide are probably compatible with breastfeeding. Chlorpropamide and tolbutamide are excreted into breast milk.77 [EL = 3] 144 . The reference guide also suggested that the amount of acarbose available for transfer to breast milk is very small because less than 2% of the acarbose dose is absorbed systemically. [EL = 2−] Another cohort study393 investigated whether glibenclamide and glipizide may be excreted into breast milk and whether breastfeeding from women taking these drugs causes infant hypoglycaemia. pioglitazone. imidapril. There was no information about captopril. Infant blood glucose was measured 5–16 days after birth. Both studies concluded that metformin was safe to use during breastfeeding. imidapril hydrochloride or ramipril.Diabetes in pregnancy metformin on the first day after caesarean section. and suggested that they are probably compatible with breastfeeding. Another five women treated with 5 mg/day of glibenclamide or glipizide starting on the first day postpartum were assessed by measuring the concentration of the drugs in maternal blood and milk. fosinopril and lisinopril have been found to be present in breast milk and they should be avoided by women who are breastfeeding. Neither glibenclamide nor glipizide were detected in breast milk and blood glucose was normal in the three babies (one glibenclamide and two glipizide) who were wholly breastfed. but it suggests that they are probably compatible with breastfeeding. The manufacturer of acarbose advises women who are breastfeeding to avoid it. Sulphonylureas have a theoretical possibility of causing hypoglycaemia in the baby.78 Angiotensin-II receptor blockers A reference guide to drugs in pregnancy and lactation reported that there were no data for the use of ARBs in women who are breastfeeding. moexipril. The manufacturers of trandolapril advise pregnant women to avoid it. leaving only three for analysis. repaglinide. but the reference guide suggested that it may produce skeletal deformities. pioglitazone and rosiglitazone advise women who are breastfeeding to avoid them. The results are not meaningful and the study is not considered further.77 [EL = 3] The British National Formulary reports that quinapril.77 The review included evidence from two small observational studies in breastfeeding women which found that metformin is excreted in milk. [EL = 3] The reference guide reported that repaglinide. [EL = 3] The reference guide reported that acarbose. No studies have investigated the use of repaglinide in breastfeeding women. There were no data for the use of lisinopril.

fluvastatin. telmisartan and valsartan have no information available so the manufacturers advise women who are breastfeeding to avoid them. isradipine. but no further information is given.77 Studies have shown that fluvastatin and pravastatin appear in breast milk. but in an amount too small to be harmful.78 Calcium-channel blockers A reference guide to drugs in pregnancy and lactation reported that there are no data for the use of the calcium-channel blockers amlodipine. eprosartan. A small cohort study found lower FBG levels 6 weeks after birth in women who breastfed than in women who bottle-fed or discontinued breastfeeding before 6 weeks.Postnatal care The British National Formulary states that olmesartan has been found to be present in breast milk and recommends that it should be avoided by women who are breastfeeding. nicardipine or nisoldipine in women who are breastfeeding. Blood glucose levels fell only for the first week postpartum.78 Evidence statement There are no high-quality studies that show that breastfeeding affects glycaemic control. but that it is probably compatible with breastfeeding. nicardipine or nisoldipine and the manufacturers advise breastfeeding women to avoid them. however breastfeeding women were advised to eat a meal or snack before feeds and the small numbers in the study limits comparison between groups. 145 . atorvastatin or rosuvastatin. A case series found a significant reduction in insulin requirements in breastfeeding women following birth. rosuvastatin and simvastatin during breastfeeding and women who are breastfeeding are advised to avoid them. Nifedipine is found in breast milk.77 [EL = 3] The British National Formulary states that verapamil has been found to be present in breast milk in an amount too small to be harmful. Felodipine is present in milk. The reference guide reported that there are limited data for the use of diltiazem. however. nifedipine. losartan. irbesartan. lercanidipine. No data were available for lacidipine or lercanidipine hydrochloride. but the study was underpowered to detect a difference in insulin requirements between breastfeeding and bottle-feeding women. There was no review for rimonabant. nimodipine and verapamil in women who are breastfeeding and it suggested that they are probably compatible with breastfeeding. There was no difference between groups in hypoglycaemic episodes. There is no information available for amlodipine. There is no information available for nimodipine. Candesartan. A small cohort study showed that insulin requirements and blood glucose levels fell in all women with diabetes following birth. the manufacturer advises breastfeeding women to avoid it. The reference guide reported that there were no data for the use of sibutramine in women who are breastfeeding and it suggests there may be toxicity to the baby. Obesity drugs A reference guide to drugs in pregnancy and lactation reported that there were no data for the use of the obesity drug orlistat in women who are breastfeeding. but suggested that they are probably compatible with breastfeeding. Isradipine may be present in breast milk and the manufacturer advises women who are breastfeeding to avoid it. The manufacturers of orlistat and sibutramine recommend that they be avoided in women who are breastfeeding. [EL = 3] The British National Formulary notes that pravastatin has been found to be present in a small amount in breast milk and should be avoided by women who are breastfeeding.78 Diltiazem has been found to be present in a significant amount in milk and although there is no evidence of harm the manufacturer advises pregnant women to avoid it unless there is no safer alternative. No data were available for simvastatin.77 [EL = 3] The British National Formulary recommends that rimonabant be avoided by women who are breastfeeding. felodipine. There is no information available for the use of atorvastatin.78 Statins A reference guide to drugs in pregnancy and lactation reported that statins are contraindicated in women who are breastfeeding. lacidipine.

8. The SPC advises that there is insufficient/limited information on the excretion of glibenclamide in human or animal breast milk. ARBs. it is the GDG’s view that women with pre-existing type 2 diabetes who are breastfeeding can resume or continue to take metformin and glibenclamide immediately following birth. Informed consent on the use of glibenclamide during lactation should be obtained and documented. Women with pre-existing type 2 diabetes who are breastfeeding can resume or continue to take metformin and glibenclamide immediately following birth but other oral hypoglycaemic agents should be avoided while breastfeeding. especially when breastfeeding. The safety of oral hypoglycaemic agents.  Glibenclamide is used in UK clinical practice in the management of diabetes in pregnancy and lactation.Diabetes in pregnancy There is limited evidence from two cohort studies and a reference guide in relation to the safety of oral hypoglycaemic agents. Women with diabetes who are breastfeeding should. There were no research recommendations relating to breastfeeding and effects on glycaemic control in women with diabetes. There is strong evidence for its effectiveness and safety. Informed consent on the use of metformin during lactation should be obtained and documented. Recommendations for breastfeeding and effects on glycaemic control Women with insulin-treated pre-existing diabetes should reduce their insulin immediately after birth and monitor their blood glucose levels carefully to establish the appropriate dose. glucose metabolism in women who have been diagnosed with gestational diabetes may return to normal. This evidence is not currently reflected in the SPC (July 2008). or there may be ongoing impaired glucose regulation (IGT or impaired fasting glycaemia) or frank diabetes (including pre-existing type 1 or type 2 diabetes that was unrecognised before pregnancy). This evidence is not currently reflected in the SPC (July 2008). Women who have been diagnosed with gestational diabetes should discontinue hypoglycaemic treatment immediately after birth. be advised to reduce their insulin dose immediately after birth and to monitor their blood glucose levels to establish the appropriate dose. calcium-channel blockers and obesity drugs in women who are breastfeeding has not been established. However. statins. There is strong evidence for its effectiveness and safety. continue to avoid any drugs for the treatment of diabetes and its complications that were discontinued for safety reasons in the preconception period. The SPC advises that metformin is contraindicated in lactation. Women with insulin-treated pre-existing diabetes should. Women with insulin-treated pre-existing diabetes who are breastfeeding should be informed that they are at increased risk of hypoglycaemia when breastfeeding and to have a meal or snack available before or during feeds. therefore. and they should be advised to have a meal or snack available before or during feeds. From evidence to recommendations Given the lack of clinical evidence in relation to the effect of breastfeeding on glycaemic control the GDG’s recommendations are based on consensus within the group on best current practice. statins. ACE inhibitors. calcium-channel blockers and obesity drugs in women who are breastfeeding. Women with insulin-treated pre-existing diabetes should be informed that they are at increased risk of hypoglycaemia in the postnatal period. ACE inhibitors. ARBs. The reference guide and the manufacturers of the drugs recommend that these preparations are avoided by women who are breastfeeding. therefore.31 One systematic review and nine additional or subsequent studies examining the likelihood of women who have been diagnosed with gestational diabetes later developing type 1 or type 2  Metformin is used in UK clinical practice in the management of diabetes in pregnancy and lactation. Women who have been diagnosed with gestational diabetes should discontinue hypoglycaemic treatment immediately after birth.2 Information and follow-up after birth Description of the evidence Gestational diabetes In the postnatal period. 146 . Women with diabetes who are breastfeeding should continue to avoid any drugs for the treatment of diabetes complications that were discontinued for safety reasons in the preconception period.

The study recommended that prospective diabetes assessment and intervention should be considered in women with gestational diabetes who are autoantibody positive.3% in the old cohort. at most.05). Asian origin and 1 hour plasma glucose were predictive of developing diabetes. The review concluded that women with higher fasting glucose levels during pregnancy may need to be tested for type 2 diabetes more often than current guidelines recommend. Both cohorts were followed up in 2002 (n = 481) with a median follow-up of 9. The 1999 WHO criteria were used for classification. Comparing the cohorts. being overweight before pregnancy (BMI 25 kg/m² or more) and IGT postpartum were statistically significant risk factors for developing diabetes (P < 0. The study found that 2. previous history of gestational diabetes. A systematic review (28 studies) examined risk factors associated with developing type 2 diabetes in women who had been diagnosed with gestational diabetes. Two RCTs and a cross-sectional study on the effect of lifestyle/educational interventions on the development of type 2 diabetes were identified. and the recurrence rate is about 75% in women with a history of insulintreated gestational diabetes (see Section 4. BMI more than 30 kg/m² and serum C-reactive protein at 9 months in 2nd to 4th quartiles were statistically significant predictors of developing diabetes. Whilst this study included a large number of women the follow-up period was. length of follow-up and criteria for diagnosis of gestational diabetes and type 2 diabetes and that this made comparison and generalisation of results difficult.395 Until 1986. whereas afterwards a 75 g OGTT was used. and that the increase in BMI in the population seemed to be the main risk factor accounting for this. n = 151) with a cohort recruited between 1987 and 1996 (new cohort. Recurrence rates for gestational diabetes vary between 30% and 84% after the index pregnancy. BMI.6% to 70% over periods from 6 weeks to 28 years. The study found that having a first-degree relative with diabetes. The epidemiological data showed that the incidence of type 2 diabetes increased most rapidly in the first 5 years after pregnancy. The study concluded that these risk factors should be taken into account when deciding follow-up care for women with gestational diabetes. The review highlighted that the included studies varied in ethnicity. 40. The study found that insulin use during pregnancy. family history of diabetes or parity. [EL = 2−] A cohort study (n = 302) undertaken in Germany examined the risk factors associated with developing diabetes after being diagnosed with gestational diabetes. duration of pregnancy.0% (58/2956) of women developed diabetes within the first 6 months postpartum. birthweight of child and number of previous pregnancies were not predictive of subsequent diabetes. Multivariate analysis found that severity of gestational diabetes. The study concluded that the incidence of diabetes in the cohort was very high and increasing. [EL = 3] 147 . There was no clear pattern for risk factors such as BMI. [EL = 3] A cohort study using routinely collected data (n = 2956) from Australia examined risk factors for developing diabetes after being diagnosed with gestational diabetes. Epidemiology: Women who have been diagnosed with gestational diabetes are likely to have gestational diabetes in future pregnancies. n = 330).397 The study used a 50 g OGTT and the WHO criteria for classification. maternal age and booking status (private or not) were not. Fasting glucose levels from OGTTs administered during pregnancy were predictive of developing type 2 diabetes after pregnancy. 6 months. A further systematic review on the effectiveness of pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people with IGT was identified.396 The study used a 75 g OGTT and the American Diabetes Association criteria for classification.9% in the new cohort had type 1 or type 2 diabetes compared with 18. a 3 hour 50 g OGTT was used.394 The studies included in the review reported rates of conversion to type 2 diabetes from 2. but that insulin treatment during pregnancy. require insulin treatment during pregnancy or who are obese. Two studies on alternatives to a 6 week OGTT for women who have been diagnosed with gestational diabetes were identified. Multiple regression analysis showed that membership of the new cohort. maternal age.8 years. fetal macrosomia. The study found that overall 40% (192) of women had type 1 or type 2 diabetes and 27% (130) had impaired glucose regulation (IGT or impaired fasting glycaemia).Postnatal care diabetes were identified. age. [EL = 2+] A cohort study (n = 753) from Denmark compared the incidence of diabetes after gestational diabetes in a cohort of women with gestational diabetes recruited between 1978 to 1985 (old cohort.1).

At 1 year follow-up 22% (50/229) of cases and 1. Multiple regression analysis found that the results of OGTT test during pregnancy was predictive for developing diabetes later. recurrent gestational diabetes and use of oral contraceptives were not predictive of later developing diabetes.6% (1/60) of controls had developed type 2 diabetes (P < 0. The study used the 2 hour 75 g OGTT.8% (5/58 and 3/58.403 The average follow-up period was 24 months. The study found that 46% (114/244) of non-pregnant women and 23% (17/73) of pregnant women had developed diabetes within the 10 year follow-up period.401 The study used the Finnish Diabetes Association classification and a 75 g OGTT. weight increase. Ten percent (43/435) of women in the case group had developed type 1 or type 2 diabetes. P < 0.Diabetes in pregnancy A cohort study (n = 278) from Hong Kong compared women with abnormal glucose tolerance test results with those with normal glucose tolerance test results. [EL = 2−] A case–control study (n = 468. women had reduced the fat intake in their diets (58 compared with 90 before pregnancy).398 The study used a 75 g OGTT and the WHO criteria for classification.001). respectively. Regression analysis showed that age. positive islet cell antibodies.001). whereas 20% (18/90) of women with insulin-treated gestational diabetes and 2% (3/132) of women with non-insulin-treated gestational diabetes had 2 hour OGTT values of more than 11.2 years versus 34.0 ml. The authors concluded that IGT outside pregnancy was a stronger predictor of developing diabetes than IGT during pregnancy.0% (56/193) of women who had been diagnosed with gestational diabetes had IGT (n = 38) or diabetes (n = 18) by 6 years followup compared with 13.05).0 years. The study found that 19 women had developed diabetes and 22 had IGT. whereas none of 52 controls had developed diabetes (P < 0. parity and not being pregnant were all statistically significant risk factors in developing diabetes. estimated fetal weight and birthweight were not associated with developing diabetes. However. but BMI.001). On average the women had gained weight after pregnancy (36 gained weight compared with 18 who lost weight) and they were not exercising as much after pregnancy as before (36 not exercising before pregnancy versus 47 not exercising after pregnancy).8–11. [EL = 2−] A retrospective case-series (n = 121) from Denmark examined lifestyle changes after pregnancy in women who had been diagnosed with gestational diabetes. This highlights the often transient nature of gestational diabetes. diabetes in a first-degree relative.0 mmol/litre. fasting blood sugar and HbA1c were all significant predictors of women with gestational diabetes developing diabetes compared to women with gestational diabetes that did not develop diabetes (P < 0. [EL = 2+] A cohort study (n = 317) from the USA compared the incidence of type 2 diabetes in Pima Indians who had IGT (75 g) and who were either pregnant or not when the test was undertaken. positive glutamic acid decarboxylase antibodies and being positive for more than one antibody were all predictive of developing diabetes. Women treated with insulin during pregnancy were more likely to develop diabetes than those who did not use insulin (P < 0. respectively) of women without gestational diabetes. Twenty-seven percent (24/90) of women with insulin-treated gestational diabetes and 17% (23/132) women with non-insulin-treated gestational diabetes had 2 hour OGTT values of 7. whereas none of the women in the control group had developed diabetes. The study found 2 hour OGTT value and HbA1c at diagnosis were associated with diabetes at 1 year.402 The study used the WHO criteria for classification and a 75 g OGTT. Using multiple regression analysis the study found that 2 hour plasma glucose.0001). [EL = 3] A case–control study (n = 70) from Sweden compared the incidence of type 2 diabetes at 15 years follow-up between women who had gestational diabetes and those who did not. macrosomia. abnormal OGTT at 6 weeks postpartum. The study found that 29. The women in the control group were significantly younger (27.399 The study found that 35% (10/28) of women with gestational diabetes had developed type 2 diabetes. insulin treatment. BMI. [EL = 2+] A case–control study (n = 870) from Finland compared the incidence of type 1 or type 2 diabetes postpartum in women who had been diagnosed with gestational diabetes (n = 435) with those who had not (n = 435).400 The study used a 2 hour 75 g OGTT and the European Association for the Study of Diabetes criteria for classification. BMI. 315 cases and 153 controls) from Sweden compared women with and without gestational diabetes for later development of diabetes. [EL = 3] 148 . The study found that age. Weight. The study concluded that better postpartum strategies for control of weight and lifestyle are needed for women who have been diagnosed with gestational diabetes.

Using a cut-off for FPG of 6. At 2.8 cases per 100 person-years.408 The study compared FPG with OGTT (as the gold standard). and it focused on prevention of diabetes rather than management of existing diabetes.3%. exercise and behaviour plus group sessions to reinforce behaviour.5 mmol/litre the sensitivity was 73.8 year follow-up the incidence of diabetes in the placebo group was 11 cases per 100 person-years. However. whereas for metformin it was 7. and for the lifestyle change programme it was 4. At 4 year follow-up the cumulative incidence was 11% in the individualised counselling group (95% CI 6% to 15%) and 23% in the standard information group (95% CI 17% to 29%).2 years. The study involved women and men with an average age of 55 years with high-risk factors for developing type 2 diabetes. [EL = 2] A retrospective diagnostic study (n = 298) from Singapore examined whether the results of an antenatal OGTT could be used to predict which of those women who had been diagnosed with gestational diabetes would go on to develop diabetes. the results may not be applicable to pregnant women. A total of 122 women had results available for analysis. and for women only (n = 2191) the figure was 36% (95% CI 16 to 51). metformin plus standard advice (n = 1073) and an intensive lifestyle change programme (n = 1079) in the prevention of development of type 2 diabetes. Lifestyle interventions seemed to be at least as effective as pharmacological treatment. age 40–65 years and IGT) compared the effect of individualised counselling (n = 265) with standard information provision (n = 257) on the prevention of diabetes. The study involved women and men with an average age of 50.9% and specificity was 70. The study showed low prevalence of physical activity that was strongly related to social support and selfefficacy.0 mmol/litre. Therefore. The groups were comparable at baseline. but not specifically women and pregnancy. The study concluded that FPG could be used to determine who should undergo an OGTT. The study shows that intensive lifestyle education reduced the incidence of type 2 diabetes.Postnatal care Lifestyle interventions: An RCT (n = 3234) undertaken in the USA of people (women and men) with elevated fasting and post-load plasma glucose concentrations compared placebo plus standard advice (n = 1082). [EL 3] A systematic review and meta-analysis407 of 17 RCTs attempted to quantify the effectiveness of pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people (women and men) with IGT.405 Cumulative incidence of diabetes in the individualised counselling group was 58% lower than in the standard information group.7%. with a mean follow-up of 3. The authors concluded that antenatal OGTT results could not be used reliably to predict postnatal OGTT results. Finally no cost data were available to determine the cost-effectiveness of the interventions. the reduction in development of diabetes for the general population is likely to be greater than those shown in the placebo group because the fact that these people were made aware of the problem is likely to have had some impact.6 years. No separate analyses for women and men were reported. At a cut-off of 4.404 The intensive lifestyle change programme involved one-to-one meetings over 24 weeks focusing on changing diet.8 cases per 100 person-years. For a 2 hour OGTT the cut-off was 10. [EL = 1+] Follow-up screening: A retrospective diagnostic study (n = 152) from the UK examined whether an FPG test at 6 weeks postpartum could be used to determine which women needed an OGTT. The study showed that lifestyle and pharmacological interventions reduced the rate of progression to type 2 diabetes. the aim being to avoid the need for a 6 week follow-up OGTT. [EL = 1+] An RCT undertaken in Finland (n = 522) in people at high risk of developing type 2 diabetes (relatives with type 2 diabetes. BMI more than 25 kg/m². the sensitivity was 100% and the specificity was 94% for identifying those who had diabetes compared to OGTT.1% and a specificity of 84.409 The study compared the antenatal OGTT results with the postnatal OGTT results. The reduction in incidence between the lifestyle change programme group and metformin group was 39% (95% CI 24 to 51). [EL = 3] 149 . [EL = 1+] A cross-sectional study406 examined postpartum patterns of physical activity and related psychosocial factors in women who had been diagnosed with gestational diabetes.5 mmol/litre with a sensitivity of 55. The trial was stopped early by the data monitoring committee due to the divergence in the placebo group.

008). Those who are found still to have impaired glucose regulation should also be offered a full assessment of their cardiovascular risk and appropriate follow-up.95) and to receive suboptimal postnatal diabetes care (53% versus 46%. Sixty-six percent (133/203) of the women who had a poor pregnancy outcome and 50% (106/211) of the women who had a good pregnancy outcome were classified as having had sub-optimal postnatal diabetes care and advice (OR 1.3). 95% CI 2. Gestational diabetes: Six weeks after delivery. P = 0. emphasising the importance of good communication between maternity and diabetes teams and that maternity staff should have good access to expert advice about glycaemic control.8. adjusted for maternal age and deprivation). adjusted for maternal age and deprivation). [EL = 3–4] The CEMACH enquiry (comparison of women with type 1 and type 2 diabetes) reported that women with type 1 diabetes were as likely to have a written plan for postnatal diabetes management as women with type 2 diabetes (87% versus 87%. lack of contact with the diabetes team and lack of contraceptive advice for women with pre-existing type 1 and type 2 diabetes postnatally. a risk that can be reduced by eating a balanced diet. 95% CI 1. P = 0. or • now has impaired glucose tolerance. and • type 2 diabetes later in life. a 75 g oral glucose tolerance test should be undertaken to determine whether the woman: • still has diabetes. P = 0. 150 . maintaining a healthy weight and increasing their physical activity levels. all women should be offered the opportunity to be reviewed by the multidisciplinary team and to discuss the future self-management of their diabetes and the implications of breastfeeding. The CEMACH enquiry described the postnatal care of women with pre-existing type 1 and type 2 diabetes.7. and offering women a followup diabetes appointment after discharge from hospital to discuss ongoing management of diabetes.2 to 2. Those who are found still to have impaired glucose regulation and those who have returned to normal should be advised that they have an increased risk of developing: • gestational diabetes in subsequent pregnancies.33 It also commented on: the need for a clear written plan for diabetes management in the woman’s medical records. The enquiry panels expressed concern specifically about the management of glycaemic control. offering information and advice about contraception and the importance of planned pregnancy before the woman is discharged from hospital.’ The NICE postnatal care guideline11 recommends that resumption of contraception should be discussed within the first week of birth.Diabetes in pregnancy Type 1 and type 2 diabetes No clinical studies were identified in relation to the information and follow-up that should be provided for women with type 1 diabetes and type 2 diabetes in the postnatal period. OR 4.33 [EL = 3–4] Existing guidance The NSF for diabetes20 recommends that services should be in place for women with pre-existing diabetes and those who have been diagnosed with gestational diabetes.4. ‘Pregestational diabetes: Following delivery.4 to 7. Women with type 1 diabetes were more likely to have postnatal contraceptive advice compared to women with type 2 diabetes (85% versus 70%. inadequate plans for care after discharge.2. They should all be offered contraceptive advice and should all receive a six-week postpartum check. The enquiry reported that 17% (31/184) of women who had poor pregnancy outcome and 13% (25/188) of women with good pregnancy outcome had no documented plan for postnatal diabetes management and 73% (280/383) had a follow-up diabetes appointment planned. They should also be given advice about the symptoms and signs of diabetes. Women who had a poor pregnancy outcome were more likely not to receive contraceptive advice before being discharged from hospital (44%) than those with a good pregnancy outcome (16%. Women who are found still to have diabetes should be managed accordingly. or • has returned to normal.

rather than an OGTT. They are also less costly than OGTTs and it is the GDG’s view that using OGTTs instead of FPG measurements would not affect outcomes. and the costs are probably the same. In recommending that women who have been diagnosed with gestational diabetes are informed that they are likely to have gestational diabetes in future pregnancies. Given that no clinical studies that evaluated the information and follow-up that should be provided for women with type 1 or type 2 diabetes in the postnatal period were identified. women who have been diagnosed with gestational diabetes should be offered lifestyle advice and follow-up to have their blood glucose tested at the 6 week postnatal check and annually thereafter. Women who have had gestational diabetes in a previous pregnancy should. therefore. be offered blood glucose testing using FPG. the GDG’s 151 . This represents a change in clinical practice that will bring a cost saving to the NHS. length of follow-up. There is evidence that women who have been diagnosed with gestational diabetes are likely to have gestational diabetes in future pregnancies. use of insulin during pregnancy. these studies were undertaken on a general population at risk of developing diabetes rather than women who had been diagnosed with gestational diabetes. the GDG is reinforcing the recommendations contained in the NSF for diabetes. From evidence to recommendations In the postnatal period. suggesting a clinical need for postnatal testing of women who have been diagnosed with gestational diabetes. However. Women who have been diagnosed with gestational diabetes should. No clinical studies that evaluated the information and follow-up that should be provided for women with type 1 diabetes and type 2 diabetes in the postnatal period were identified. therefore.3). be offered early self-monitoring of blood glucose or OGTT. Recurrence rates of gestational diabetes are between 30% and 84%. therefore. The studies highlight various risk factors or identifiers for developing diabetes after having had gestational diabetes. However. and results of OGTTs during pregnancy. There is evidence that lifestyle/education interventions are effective for people with IGT to prevent progression to type 2 diabetes and. with recurrence rates in women with a history of insulin-treated gestational diabetes being about 75%. Women who have been diagnosed with gestational diabetes should. or there may be ongoing impaired glucose regulation (IGT or impaired fasting glycaemia) or frank diabetes (including pre-existing type 1 or type 2 diabetes that was unrecognised before pregnancy). There is no clinical evidence to support early self-monitoring of blood glucose (for 1 week) over OGTT in future pregnancies. Women who have been diagnosed with gestational diabetes are likely to develop type 2 diabetes postnatally and so they should be informed of the symptoms of hyperglycaemia. Results from a systematic review of epidemiological studies and eight additional or subsequent studies show increasing cumulative incidence of type 2 diabetes in the postnatal period in women who have been diagnosed with gestational diabetes. the main ones being obesity. therefore. glucose metabolism in women who have been diagnosed with gestational diabetes may return to normal. be offered blood glucose testing before they are discharged from hospital to exclude persisting hyperglycaemia. long-term follow-up would be needed to determine the effectiveness of the programmes. Two diagnostic studies showed that follow-up of women with gestational diabetes was required to accurately identify ongoing disruption of glucose metabolism. high attrition rates at follow-up and differing ethnic and cultural populations. the studies are limited by the variation in data recorded. and a further OGTT if the results are normal (see Section 4. There is evidence from a diagnostic study that FPG measurements have high sensitivity and specificity compared with OGTTs (the gold standard). A systematic review and two RCTs of lifestyle/education interventions showed that the risk of developing type 2 diabetes could be reduced by either lifestyle or pharmacological interventions.Postnatal care Evidence statement Evidence shows that women who have been diagnosed with gestational diabetes are likely to have gestational diabetes in future pregnancies. or vice versa. In addition.

Diabetes in pregnancy recommendation is based on the consensus view of the group. Randomised controlled trials are needed to determine the clinical and cost-effectiveness of such treatments compared to diet and exercise. The phrase ‘women who have been diagnosed with gestational diabetes’ is used in the recommendations contained in this section to highlight the fact that the gestational diabetes may have resolved immediately postpartum. Women with pre-existing type 1 or type 2 diabetes should. Women who were diagnosed with gestational diabetes (including those with ongoing impaired glucose regulation) should be informed about the risks of gestational diabetes in future pregnancies and they should be offered screening (OGTT or fasting plasma glucose) for diabetes when planning future pregnancies. diet and exercise) and offered a fasting plasma glucose measurement (but not an OGTT) at the 6 week postnatal check and annually thereafter. Women with diabetes should be reminded of the importance of contraception and the need for preconception care when planning future pregnancies. Women who were diagnosed with gestational diabetes (including those with ongoing impaired glucose regulation) should be offered early self-monitoring of blood glucose or an OGTT in future pregnancies. therefore. Research recommendations for information and follow-up after birth Are there suitable long-term pharmacological interventions to be recommended postnatally for women who have been diagnosed with gestational diabetes to prevent the onset of type 2 diabetes? Why this is important Oral hypoglycaemic agents such rosiglitazone and metformin offer the possibility of pharmacological treatment for prevention of progression to type 2 diabetes in women who have been diagnosed with gestational diabetes. Recommendations for information and follow-up after birth Women with pre-existing diabetes should be referred back to their routine diabetes care arrangements. Women who were diagnosed with gestational diabetes should be offered lifestyle advice (including weight control. This care should include consideration of the issues identified in relation to preconception care. As yet there have been no clinical studies to investigate the effectiveness of oral hypoglycaemic agents in this context. Women who were diagnosed with gestational diabetes should be reminded of the symptoms of hyperglycaemia. including the importance of contraception and planning future pregnancies (see Chapter 3). be referred back to their routine follow-up arrangements with the diabetes care team. A subsequent OGTT should be offered if the test results in early pregnancy are normal. Women who were diagnosed with gestational diabetes should have their blood glucose tested to exclude persisting hyperglycaemia before they are transferred to community care. 152 .

Novo Nordisk.Appendix A Declarations of interest This appendix includes all interests declared on or before 28 February 2008. Novo Nordisk. the Primary Care Diabetes Society receives sponsorship for educational programmes from Eli Lilly. Warwickshire. Pfizer. Sanofi Aventis.1 Guideline development group members Dominique Acolet No interests declared Lynne Carney Personal non-pecuniary interests: Speaker at Welsh CEMACH conference Non-current interests – planned: Teacher on specialist antenatal course for women with diabetes Anne Dornhorst Personal pecuniary interests – specific: Consultancy for GlaxoSmithKline. GlaxoSmithKline. conference expenses and/or lecture fees from Aventis. UK principal investigator for PREDICTIVE post-marketing surveillance study for treatment of type 1 and type 2 diabetes using insulin detemir and insulin aspart funded by Novo Nordisk. conference expenses and/or lecture fees from Bristol-Myers Squibb. Sanofi Aventis and Takeda Personal non-pecuniary interests: Medical adviser to Warwick Diabetes Care. University of Warwick. Sanofi Aventis. Lifescan. Novo Nordisk and Takeda. Servier and Takeda Non-current interests – previous: Consultancy. Colgate-Palmolive. Merck Sharp & Dohme Limited. Sanofi Aventis and Servier. Novo Nordisk. Roche. Novartis. Servier and Takeda. Roche Diagnostics. Merck Sharp & Dohme Limited. Warwick Diabetes Care received start-up 153 . GlaxoSmithKline. Novo Nordisk and Servier Personal non-pecuniary interests: Officer of the Royal College of Physicians. Roche Diagnostics. Merck Pharma. Osaki. Eli Lilly. Member of the Working Lives intercollegiate committee Non-personal pecuniary interests – specific: Hospital department receives funding from Novo Nordisk in connection with the PREDICTIVE study and insulin detemir in pregnancy study Robert Fraser No interests declared Roger Gadsby Personal pecuniary interests – specific: Adviser to Bristol-Myers Squibb. Honorary Treasurer of the Primary Care Diabetes Society Non-personal pecuniary interests – specific: Warwick Diabetes Care receives sponsorship for educational programmes from the British In Vitro Diagnostics Association (BIVDA). Nuneaton. Merck. A. GlaxoSmithKline. GlaxoSmithKline. Pfizer. Merck Pharma. Novo Nordisk. Chairman of Trustees of Pregnancy Sickness Support.

Bayer. Roche and Takeda Non-personal pecuniary interests – specific: Investigator for Softsense blood glucose meter in pregnancy study funded by Abbott Laboratories and insulin aspart and insulin detemir in pregnancy studies funded by Novo Nordisk. funded by Novo Nordisk to work on an outof-hours helpline and to attend related update meetings Personal non-pecuniary interests: Member of Diabetes UK and Royal College of Nursing. Roche Diagnostics and Sanofi Aventis. Eli Lilly. Novo Nordisk. Owen Mumford. GlaxoSmithKline. participation in CEMACH meetings. lecture fees and educational grants from Astra-Zeneca. Merck Sharp & Dohme Limited. Lifescan. Bayer. Eli Lilly. Department Trust fund receives funding to support attendance at conferences. courses. Lifescan. Menarini. Diabetes UK. Novartis. Eli Lilly. Becton and Dickenson. Pfizer and Takeda Jane Hawdon Personal non-pecuniary interests: Chair of neonatal working group for the CEMACH Diabetes in Pregnancy Enquiry. insulin detemir study funded by Novo Nordisk Saiyyidah Zaidi No interests declared 154 .Diabetes in pregnancy sponsorship from Aventis. Novo Nordisk. Becton and Dickenson. BLISS and CEMACH Richard Holt Personal pecuniary interests – specific: Investigator for insulin aspart and insulin detemir in pregnancy studies funded by Novo Nordisk. conference expenses and/or lecture fees from Eli Lilly and GlaxoSmithKline Personal pecuniary interests – non-specific: Consultancy. adviser and speaker for Baby Friendly Initiative. Novo Nordisk. Bristol-Myers Squibb. Menarini. GlaxoSmithKline. study days. Eli Lilly. Roche Diagnostics. attended a meeting of the Management of Diabetes for Excellence (MODEL) group Non-personal pecuniary interests – specific: Adviser on patient education literature for Eli Lilly. Sanofi Aventis and the Centre for Pharmacy Postgraduate Education. Novo Nordisk. research funding from GlaxoSmithKline to examine the role of insulin resistance in gestational diabetes Personal non-pecuniary interests: Chair of the Professional Advisory Council of Diabetes UK Ann Parker No interests declared Nickey Tomkins No interests declared Stephen Walkinshaw No interests declared Jackie Webb Personal pecuniary interests – specific: Conference/meeting expenses and/or lecture fees from Abbot Diabetes Care. meetings and patient-support events and meetings from Abbot Diabetes Care. adviser on GlucoGel for British BioCell. University of Manchester. GlaxoSmithKline. Lifescan. GlaxoSmithKline.

Novartis (dipeptidyl peptidase 4 (DPP-4) inhibitor). Merck Sharp & Dohme Limited (cardiovascular preparations).2 NCC-WCH staff and contractors Paula Broughton-Palmer No interests declared Michael Corkett No interests declared Anthony Danso-Appiah No interests declared Paul Jacklin Non-current interests – planned: Commissioned by Continence-UK to write an article on the health economics of continence care Lorelei Jones No interests declared Moira Mugglestone No interests declared Jeffrey Round No interests declared Anuradha Sekhri No interests declared A.3 External advisers Anita Holdcroft Personal pecuniary interests – non-specific: Consultancy for special opioid advisory group. sponsorship from various pharmaceutical companies for organising meetings of the Royal Society of Medicine’s Section of Anaesthesia Jo Modder No interests declared Peter Scanlon No interests declared Roy Taylor Personal pecuniary interests – non-specific: Adviser to Eli Lilly (protein kinase C (PKC) inhibitor). Napp pharmaceuticals Non-personal pecuniary interests – non-specific: Research funding from Cephalon.Appendix A A. conference expenses from Eli Lilly and Merck Sharp & Dohme Limited Non-personal pecuniary interests – non-specific: Funding from Servier to support production of educational tools (book and website) on retinal screening in diabetes 155 .

and (ii) women with diabetes who are planning a pregnancy? Gestational diabetes 11. Intrapartum care 24. 16. What are the barriers to uptake of preconception care? 10. monitoring and intervention for gestational diabetes improve outcomes in mothers and babies? 14. Which women are at high risk of gestational diabetes? 12. How cost-effective are self-management programmes for women with diabetes who are planning a pregnancy? 7. dietary supplements. What special considerations in relation to spontaneous or planned preterm birth are appropriate for women with diabetes? 15. When and by what methods should fetal growth and wellbeing be monitored? 22.Appendix B Clinical questions Preconception care 1. What information should be offered in relation to the importance of planning a pregnancy and the role of contraception? 3. Which medications for diabetic complications are suitable for use during pregnancy and which should be discontinued? 9. Does intervening have any implications for future pregnancies and births? 26. What information should be offered in relation to diet. What information should be offered in relation to outcomes and risks for the mother and the baby? 2. How should gestational diabetes be diagnosed? 13. Does diagnosis. Does intervening in the timing and mode of birth improve outcomes for women with diabetes and their babies? 25. 19. What is cost-effective treatment for gestational diabetes? Antenatal care What are the target ranges for blood glucose during pregnancy? How should blood glucose and ketones be monitored during pregnancy? What special considerations apply to the management of diabetes during pregnancy? When and how often should women be offered retinal assessment? When and by what method should women be offered renal assessment? When and by what method should women be offered screening for congenital malformations and counselling? 21. body weight and exercise? 4. What special considerations in relation to analgesia and anaesthesia are appropriate for women with diabetes? 27. Which medications for diabetes are suitable for use during pregnancy and which should be discontinued? 8. How should blood glucose and ketones be monitored in the preconception period? 6. When should information be offered to: (i) women of reproductive age with diabetes. 18. 17. What timetable of antenatal appointments should be offered to women with diabetes? 23. What are the target ranges for blood glucose in the preconception period? 5. How should glycaemic control be monitored and maintained during labour and birth? 156 . 20.

How does breastfeeding affect glycaemic control? 32. What information and follow-up should be offered to women with type 1 and type 2 diabetes after birth? 157 .Appendix B Neonatal care 28. What initial assessment should babies undergo? Postnatal period 31. What information and follow-up should be offered to women with gestational diabetes after birth? 33. What are the criteria for admission to intensive/special care? 29. How should neonatal hypoglycaemia be prevented and treated? 30.

18 Examples of structured education programmes available in the UK are DAFNE69 for people with type 1 diabetes and DESMOND and X-PERT for people with type 2 diabetes. the study also noted that the assumption of full adherence to the preconception care and advice programme may have been a limitation of the analysis. although antenatal care would.Appendix C Cost-effectiveness of self-management programmes for women with diabetes who are planning a pregnancy C. Therefore. The study concluded that preconception care and advice would yield cost savings. it seems likely that there would be diminishing returns. expert opinion and surveys of medical care were used to estimate the costs and clinical consequences of preconception care and advice compared with ‘doing nothing’. Improvement in glycaemic control is also an important putative benefit of preconception care and advice. good glycaemic control has benefits over and above those associated with good glycaemic control outside pregnancy because of the adverse maternal and neonatal outcomes associated with poor glycaemic control in the periconceptional period and pregnancy (see Sections 3.86 as a result of fewer births. with further improvement possible. of course. An economic evaluation has suggested that structured education programmes for people with pre-existing diabetes are cost-effective in the UK setting. Indeed. The evidence suggests that such programmes lead to improved glycaemic control. generally biased against preconception care and advice. Furthermore. with each $1 spent on preconception care and advice realising a saving of $1. the study used a decision-analytic approach to determine whether. none of which fundamentally altered the results. 158 . and fewer adverse maternal and neonatal outcomes. it cannot be assumed that the effect size would be the same as preconception care and advice in the absence of a structured education programme. a de novo health economic model was developed for this guideline.410 Although not explicitly described as such. However. lower antenatal care costs arising from better glycaemic control. However. To the extent that there is further improvement in glycaemic control with preconception care and advice. as a result of averted complications.1 Introduction A review of the health economics literature identified a single study from the USA addressing the cost-effectiveness of preconception care and advice for women with pre-existing diabetes.1). additional clinical studies and a meta-analysis which included more recent data121 have been published since the cost-effectiveness study. The study is quite dated and therefore the usual caveats about the generalisability of costs from one healthcare setting to another are even more important than normal. but at a lower rate. Finally. be provided in the event of a pregnancy. the study reported that conservative estimates had been used when there was uncertainty with regard to parameter values and that their assumptions were. it is unlikely that studies of preconception care and advice have disentangled whether there are any additional improvements in relation to glycaemic control over and above those which would be achieved with a structured education programme alone. Doing nothing in this case meant no preconception care and advice. For women with diabetes who are planning a pregnancy and those who are already pregnant. The authors reported a number of sensitivity analyses.411 This is consistent with existing NICE guidance on self-management of diabetes. The study reported that a mixture of literature review. therefore. changes to parameter values undertaken as part of the sensitivity analysis may not have been quite as conservative as suggested by the authors of the study. Furthermore.4 and 5. the additional costs of preconception care and advice yielded net savings compared with no preconception care and advice.

3 and 5. advice on contraception may also improve outcomes by increasing the number of pregnancies in women with diabetes that are associated with good glycaemic control. C. there are other benefits of preconception care and advice which are specific to diabetes in pregnancy which by themselves may make its provision clinically effective and possibly costeffective.2 and C.6). Furthermore.1. C. As shown in the decision tree.Appendix C However. The model assumes that there are administration costs in just offering a preconception care and advice Figure C. In addition to the costs of the advice itself. A decision tree was developed for the guideline in Microsoft Excel® and also. It is assumed that women who are offered and adhere to preconception care and advice have a lower major congenital malformation rate than pregnant women in the no preconception care and advice arm. The effectiveness of preconception care and advice is measured in terms of the number of major congenital malformations averted.1 Preconception care and advice (PCA) versus no preconception care and advice decision tree showing major congenital malformation (CM) rates resulting from pregnancies in women with diabetes 159 . for validation purposes.3. For example. in TreeAge Pro 2006® (see Figure C. Considerable uncertainty surrounds the inputs of this model (see below for details) and therefore the results and sensitivity analysis are both undertaken to address the ‘what if’ in terms of thresholds for cost-effectiveness. advice on preconception folic acid is particularly important given the elevated risk of neural tube defects in babies of women with diabetes (see Sections 3. It is additionally assumed that those women in the preconception care and advice arm who either decline preconception care and advice or do not adhere to advice will have a congenital malformation rate equivalent to women with diabetes in the no preconception care and advice arm.2 Model parameters The parameter values used in the baseline model are shown in Tables C. the model also takes into account ‘downstream savings’ from averted congenital malformations. it is assumed that a proportion of women are infertile and therefore do not benefit from preconception care and advice even if they accept an offer of such advice and adhere to it.1).

Diabetes in pregnancy service to women with diabetes who are planning a pregnancy and therefore this is included as a cost parameter. transposition of the great arteries.9 GDG estimate GDG estimate Elixhauser et al.1 Costs (using 2006 prices) Value £10 Source GDG estimate NICE Technology Appraisal 60. renal agenesis. hydrocephalus. Table C. (2001)121 Table C. Table C.021 Ray et al.5 0. The costing of such a wide range of congenital malformations is methodologically complex and time consuming.8 0. The published cost-effectiveness study described above410 suggested that preconception care and advice would result in lower antenatal costs due to improved glycaemic control during pregnancy. As far as we are aware. The known limitations of these data are addressed by sensitivity analysis. Therefore. Anencephaly. However. Clearly there are limitations using these dated US data. equivalent UK costings do not exist. anal/rectal atresia.000 0 Source NICE guidelines manual (2007)23 A conservative baseline assumption Notes Resource item Willingness to pay for a QALY QALY gain from averted major congenital malformation 160 . However.3 Quality-adjusted life years Value £20. with the resources available for this guideline it was not possible to generate our own cost estimates based on UK NHS data. it is assumed that this cost can be limited to the population of concern (women with diabetes who are planning a pregnancy).2 Variable Probabilities Value Source 0. tetralogy of Fallot.000 Elixhauser et al. (1993)410 c A price inflation index based on changes to the price of goods and services supplied to the healthcare sector. but at baseline it conservatively assumes that preconception care and advice does not yield any cost saving in this respect compared with the no preconception care and advice alternative. Diabetes (types 1 and 2) – patient education models (2003)18 Notes Administration cost in offering a preconception care and advice service 2003 cost of £545 but updated for inflation using the Hospital and Community Health Services (HCHS)a Index At baseline it is conservatively assumed that preconception care and advice does not result in lower antenatal costs Weighted cost of major congenital malformations using an exchange rate of £1 to $2 and the HCHS index to update for inflationc Resource item Offer preconception care and advice Preconception care and £615 advice Additional costs of £0 antenatal care with poor glycaemic control Cost of a major congenital malformationb a b £81. In a similar conservative vein. spina bifida. The model structure allows this consideration to be factored in. this could be addressed in sensitivity analysis by assuming higher offer costs. caudal regression. (2001)121 Major congenital malformation rate (no preconception care and advice) 0.065 Ray et al. It does not assume that the offer is made to all women with diabetes who are of childbearing age regardless of whether they are actively planning a pregnancy. (1993)410 Decline preconception care and advice Adhere to preconception care and advice Fertile Major congenital malformation rate (preconception care and advice) 0. no QALY gain is attached to averted major congenital malformations at baseline although the results are presented to show a minimum number of QALYs per congenital malformation averted that would be needed for cost-effectiveness given a particular incremental cost of preconception care and advice. coarctation of aorta.

being cheaper and more effective than no preconception care and advice. where one parameter value was varied while holding all other parameter values constant. thresholds for cost-effectiveness were calculated for scenarios where sensitivity analysis indicated that preconception care and advice may not be the dominant strategy.2 Net cost of preconception care and advice. At one end of the spectrum there could be a group session with a diabetes specialist nurse. but a more resource-intensive model might involve a one-to-one consultation with a multidisciplinary team. Additionally. the model shows a cost saving of almost £1 million and 16 averted major congenital malformations. This sensitivity analysis was primarily undertaken on a one-way basis. This gives an indication as to whether uncertainty surrounding the exact value of the parameter is likely to have an important bearing on the model’s conclusions.000 per QALY where preconception care and advice involved incremental costs relative to no preconception care and advice. Table C.4). varying costs of preconception care and advice 161 . There are potentially many different models of preconception care and advice.540 15. In a population of 1000 women with diabetes who are planning pregnancy. There is no necessity to estimate a QALY gain to establish cost-effectiveness as preconception care and advice dominates. This involved calculating the QALY gain which would be needed to satisfy a cost-effectiveness threshold of £20. Figure C.4 Baseline results in a population of 1000 women with diabetes who are planning pregnancy Item Net costs of preconception care and advice Major congenital malformations averted Incremental cost-effectiveness ratio (ICER) Value −£965.84 Preconception care and advice dominates QALY gain needed per major congenital malformation averted N/A C.3 Results The baseline results suggest that preconception care and advice is cost-effective (see Table C.2 shows the effect of varying the cost of preconception care and advice between £50 Figure C.Appendix C C.4 Sensitivity analysis Considerable uncertainty surrounds the data inputs of the model and therefore sensitivity analysis was used to assess how robust the baseline conclusions would be given different assumptions.

If a major congenital malformation cost £5.005 then it seems likely that preconception care and advice will be cost-effective.021 (absolute difference 0.044) and 0.121 As long as the absolute difference is at least 0. varying costs of a major congenital malformation 162 .000 and £200.000. The effect of varying the assumption about the effectiveness of preconception care and advice is shown in Figure C.064 (absolute difference 0. Given the usually large impact of a major congenital malformation on lifetime health. There are a number of reasons why this may not accurately reflect costs to the NHS: • the healthcare system in the USA differs markedly from that in the UK • even in the context of the USA. The clinical effectiveness of the intervention is given by the absolute difference in the major congenital malformation rate between preconception care and advice and the no preconception care and advice alternative. the figures are presented as fairly ‘broad brush’ estimates • the original study is quite dated and treatments may have changed.75 per major congenital malformation averted would be required for cost-effectiveness.000 per woman.000 as part of a one-way sensitivity analysis. then a QALY gain of 0. it can be reasonably assumed that the QALY gain would be sufficient for cost-effectiveness in such a scenario. Preconception care and advice is dominant as long as the major congenital malformation rate with preconception care and advice is no more than 0.005 then at least five more QALYs would be needed per averted major congenital malformation and it cannot necessarily be assumed a priori that this would be the case. The results suggest that preconception care and advice would be cost saving and hence dominant even at a cost of £2.4 shows the situation where the major congenital malformation rate for no preconception care and advice is held constant at 0. This shows that preconception care and advice would be cost saving as long as a major congenital malformation cost more than £20. If the absolute difference is less than 0.001).3 Net cost of preconception care and advice.065 whilst the major congenital malformation rate for preconception care and advice is varied between the baseline 0. Figure C.Diabetes in pregnancy and £2.000.3 shows the impact of varying the cost of a major congenital malformation between £5. which is considerably less than that estimated by a recent meta-analysis.054 (absolute difference of at least 0. Another important source of uncertainty concerns the cost of a major congenital malformation. The baseline estimate was derived from a study conducted in the healthcare setting in the USA. Figure C.044.4.011). Figure C.

varying rates of major congenital malformations with preconception care and advice Figure C.5 Net cost of preconception care and advice.Appendix C Figure C. varying rates of attendance and adherence with preconception care and advice 163 .4 Net cost of preconception care and advice.

13 (the absolute difference between the upper 95% confidence limit reported in the meta-analysis for the major congenital malformation rate with preconception care and advice (worst case) and the lower 95% confidence limit for the major congenital malformation rate with no preconception care and advice (best case).121 The results of the meta-analysis need to be treated with a degree of caution because of the heterogeneous nature of the individual studies included and because of systematic differences reported between women with diabetes who attended preconception care and advice and those who did not. However. It has been suggested that the availability of a preconception clinic separates women with diabetes into two groups. and have a congenital abnormality rate of 7. If a woman’s glycaemic control has improved as a result of structured education then the scope for preconception care and advice to achieve further improvement may be limited. for various reasons. the effectiveness of preconception care and advice may be diluted when it is offered additionally to structured education programmes rather than as a stand-alone intervention. C. This is particularly important as there is likely to be considerable variation in cost between different models of preconception care and advice and it is important to know what incremental benefits are achieved by more resourceintensive forms of the intervention.005) is necessary for preconception care and advice to be considered cost-effective.5–10. the impact of changing the assumptions about attendance and adherence was assessed by comparing the worst-case scenario (no attendance with zero adherence) with the best-case scenario (full attendance and adherence with the programme).5 shows the impact of increasing attendance and adherence in equal proportions between the worst-case and bestcase scenarios. The threshold sensitivity analysis undertaken using the model suggests that only a relatively small reduction in the major congenital malformation rate (as little as 0.121 Preconception care and advice of some sort does. smokers were more prevalent in the no preconception care and advice group (30.6%) and confounding could plausibly be responsible for at least some of the observed effect. For example. 164 . If less resource-intensive preconception care and advice is almost as effective then the incremental cost-effectiveness ratio for some preconception care and advice interventions is likely to be unacceptably high.Diabetes in pregnancy Finally. book late. seem to be justified on economic grounds. This threshold for cost-effectiveness is much less than the absolute difference of 0.044 suggested in the published meta-analysis.5 Discussion The baseline effectiveness of preconception care and advice in terms of major congenital malformations averted was based on results presented in a recent meta-analysis.9%.412 Furthermore. therefore. one containing highly motivated women with well-controlled diabetes who attend and have a low rate of congenital abnormalities and the other containing women who. Initially. The published meta-analysis suggested that the preconception care and advice interventions included were heterogeneous and this raises important questions from an economic perspective in terms of what is the ‘best’ or most cost-effective form of preconception care and advice. as attendance and adherence increase the net cost of preconception care and advice increases slightly. then preconception care and advice is dominant. Figure C. producing a net cost saving in addition to the reduction in major congenital malformations.2% versus 19. when attendance and adherence reach a level of 21%. have worse glycaemic control. This is because the costs of providing preconception care and advice are not fully offset by reduced costs of congenital malformations at low adherence rates. and also less than 0.

from markets. The strategies were no screening. The authors concluded that. fetal death after 3 months of gestation or previous macrosomia) were given a nonfasting 50 g OGTT. previous macrosomia. as this may overestimate costs. pre-eclampsia. All strategies were compared with a baseline of no screening for each outcome. perinatal mortality or hypertensive disorder.1 Systematic review of screening A systematic search of the literature identified 337 studies potentially related to the clinical question. prematurity. a 75 g OGTT. 165 . respectively).31. such as caesarean section.ft. No incremental analysis was reported. 33 articles were retrieved for further appraisal and eight have been included in this section of the review. previous history of gestational diabetes. A retrospective study conducted in Italy442 examined the costs and outcomes for two groups of women. BMI greater than 27 kg/m². References were given for clinical and cost parameters but no specific details of these were reported. The authors concluded that the sequential testing strategy was cost-effective. a 100 g OGTT and a sequential test (50 g GCT followed by a 100 g OGTT). The authors recommend strategy one.1. family history of diabetes mellitus. age over 30 years and body mass). which could limit its applicability for decision making in an NHS setting. Six papers were identified that examined the cost-effectiveness of screening for gestational diabetes.1. women deemed to be at higher risk of gestational diabetes based on a series of risk factors (family history of diabetes in a first-degree relative. In strategy two all women were given the 50 g OGTT and in strategy three all women were given a 75 g OGTT. based on its favourable ICER for preventing perinatal mortality (€7870.400 for strategies two and three. D.2 Screening for gestational diabetes A cost–utility analysis443 examined four screening strategies for gestational diabetes. Two additional papers were identified that considered the cost-effectiveness of screening for and treatment of gestational diabetes.1 Screening and treatment of gestational diabetes A study conducted in France441 examined three strategies for screening for gestational diabetes using a decision analysis model. Under strategy one. compared with €8660 and €29. All women that tested positive in either screening group underwent a 100 g OGTT. After reviewing the abstracts.com/ft/markets/currencies. based on the savings from downstream interventions associated with untreated gestational diabetes. age over 35 years.Appendix D Cost-effectiveness of screening. Data on costs were collected through a prospective study of 120 pregnancies and clinical data were taken from a review of published literature. although in a high-prevalence population the 100 g OGTT may be an alternative cost-effective screening strategy. screening the population of high-risk pregnant women using the 50 g OGTT. Incremental analysis was reported in terms of cost per additional case prevented of macrosomia. Universal screening was found to be more costly than the selective screening approach per case of gestational diabetes diagnosed (€424 and €406. screening in some form was justified. diagnosis and treatment for gestational diabetes D. respectively) and that treatment cost €366. No detail was provided on what components comprised the total cost of each strategy and no unit  Exchange rate of £1 = €1. The first group had universal screening using a 50 g GCT while the second were screened based on the presence of given risk factors (history of gestational diabetes. The study was conducted from a societal perspective. D.asp on 28 February 2008.

universal GCT with 7. The authors made no conclusion on the cost-effectiveness of either approach. Six screening strategies were considered: universal fasting plasma glucose (FPG). the cost-effectiveness of treatment will partly be determined by the ability to identify patients for treatment via screening in a costeffective fashion. with a cost per case diagnosed of $80.  Exchange rate of £1 = $1.2 Introduction to the model The recently published ACHOIS trial demonstrated potential benefit of treatment for mild gestational diabetes. Four studies reported in US dollars estimated the cost per case detected of gestational diabetes. The cost per case of gestational diabetes identified by a 50 g oral glucose screening test was $114 at a cut-off of 7.444– 447 One study444 examined the cost per case diagnosed of six different strategies. The number of cases of gestational diabetes diagnosed did not differ between groups. This better illustrates the cost-effectiveness of different treatment alternatives. 166 .2 mmol/litre cutoff in women aged over 25 years. while clinical effectiveness is a necessary condition for costeffectiveness it is not sufficient. Similarly.98.665. a separate cost-minimisation analysis of the various treatment options is also presented. treatment requires identification of those affected by gestational diabetes using some screening/diagnostic strategy which further reduces scarce resources available to other NHS patients.ft. Sources for utility estimates were not provided. the cost-effectiveness of screening.asp on 28 February 2008. Therefore. One study from the UK124 examined the cost per case of gestational diabetes detected. Screening was with a 50 g GCT followed by a OGTT for women with greater than 150 mg/100 ml. with maternal and infant outcomes reported separately.56. The FPG detected an additional 6009 cases at a cost of £489 per additional case detected when compared with GCT. under the assumption of equivalent effectiveness. universal GCT with 8. where the decision to screen for cases and treat has been accepted on economic grounds. In a second study445 the authors examined the cost per case diagnosed using different thresholds for the diagnosis of gestational diabetes in a high-risk population. Women were stratified into high-. However. D. from markets. intermediate.3 mmol/litre. Therefore.com/ft/markets/currencies. Resources have competing uses and showing that resources yield a benefit does not demonstrate that an even greater benefit could not be produced if those resources were deployed in an alternative use. in addition to this single model incorporating both screening and treatment. No incremental analysis was reported. Incremental analysis was not reported. Group 1 had historical or clinical risk factors for gestational diabetes and Group 2 were offered routine screening.153 However. GCT with 8. and universal OGTT. The cost per case diagnosed of the testing programme was $329. the results of this study cannot be generalised to an NHS setting. GCT with 8. A strategy of universal OGTT was predicted to detect an additional 1493 cases compared with the universal FPG.2 mmol/litre and $106 at a cut-off of 8. A third study446 examined the cost per case diagnosed of gestational diabetes in two groups of women. Given these drawbacks. The authors recommended the use of a universal FPG or giving a GCT to those over age 25 years and with risk factors. diagnosis and treatment for gestational diabetes was developed jointly by the diabetes in pregnancy and antenatal care GDGs to enable them to make recommendations on this area of care for pregnant women. As a result. The authors recommended screening women aged over 25 years using a 50 g 1 hour glucose screening test. the cost-effectiveness of screening is predicated on an efficacious treatment which gives an acceptable cost per effect given the finite resources available.Diabetes in pregnancy costs were reported.2 mmol/litre cut-off. Furthermore.8 mmol/litre cut-off. a single cost-effectiveness model addressing screening.2 mmol/litre cut-off in women aged over 25 years and risk factors. at a cost per additional case detected of £4. The authors recommended universal screening in a high-prevalence population such as theirs.and low-risk groups based on American Diabetes Association (ADA) criteria. diagnosis and treatment for gestational diabetes are highly interdependent. Incremental analysis was undertaken and outcomes reported in qualityadjusted life years (QALYs). A fourth study447 was conducted in Iran and reported in US dollars.

Then.1).2. for diagrammatic simplicity.1 The basic decision tree structure.Appendix D D. Figure D. In this approach. As ACHOIS was limited to those with ‘mild’ gestational diabetes. Data from the ACHOIS intervention group were used to estimate the outcomes and associated costs of treating true positives. by considering the probabilities of different scenarios under each decision. the sub-tree for those with gestational diabetes who are undetected on screening is the same as the sub-tree for women with gestational diabetes who are not screened. competing alternatives represent the decisions. this cost-effectiveness model can be represented as the decision to screen and treat patients identified with gestational diabetes versus no screening. It is also necessary to consider the cost of providing treatment to women falsely diagnosed with gestational diabetes (false positives). However. At its most basic.1 The decision tree The model utilises a decision-analytic approach. The outcomes for women without gestational diabetes (true negatives and false positives) in the screening arms were not considered as the perinatal outcomes for these pregnancies do not differ from those in the population of otherwise healthy pregnant women. given an initial decision to screen there is then the decision of how to screen.3 for the treatment sub-tree. which was the recommendation of the previous antenatal care guideline (Figure D. The various screening options that have been considered in this model are described in the next section. drawing on best available evidence. [+] denotes that the tree is truncated – see Figure D.1 the decision. The outcomes and associated costs of false negatives were estimated from the routine care group in ACHOIS. Dx = diagnose 167 . In Figure D. is depicted as screen versus no screen. the costs and effects may be an underestimate of the true costs and effects in the population under consideration. the expected costs and effects of each decision can be computed and compared.

448 168 . All screening methods. The number of individuals diagnosed correctly is determined by the accuracy of the diagnostic test applied (sensitivity and specificity) and by the prevalence of the condition in the population being tested. Not all possible strategies have been considered – particularly where they are clinically inappropriate. It is not possible to consider every possible potential outcome in a model and it is important to focus on those with the greatest relevance in answering the question at hand. screening blood tests and universal diagnostic tests. based on grounds of cost and clinical effectiveness. Limitations in the data are discussed in greater detail later in this appendix. have been considered in isolation.3. This may overestimate the number of pregnancies in this group. for example treating patients based on the presence of a risk factor alone. • It has not been possible to establish an accurate fertility rate in some population subgroups.Diabetes in pregnancy The key outputs of each screening strategy are the costs of screening and treating women and the number of women accurately diagnosed with gestational diabetes.2 for details) and is assumed to be 100% sensitive and specific. Combinations of these tests have then been considered.3 Screening strategies Table D. The treatment and outcome sub-trees are identical for each screening strategy in this model but the costs and effects will vary according to the numbers diagnosed as having gestational diabetes or not. When developing a decision analysis model it is necessary to make simplifying assumptions to highlight what the important elements of the model might be and to reduce the complexity of the model. It is therefore assumed that the fertility rate among women with a high BMI is the same as the rate among women with a BMI in the normal range.1 Screening strategy assumptions Decision analysis is used to help us make decisions about the best treatment or intervention to use. see Table D. Some strategies have been excluded from further analysis after preliminary analysis showed them to be dominated by alternative strategies. D. There are four possible outcomes when applying a diagnostic test: • • • • true positive – the patient is diagnosed as positive and has the condition/disease false positive – the patient is diagnosed as positive but does not have the condition/disease true negative – the patient is not diagnosed with the condition/disease and does not have it false negative – the patient is not diagnosed with the condition/disease but does in fact have it. as high BMI is associated with fertility problems.5) BMI ≥ 27 kg/m² (high BMI) family history of diabetes. Screening blood tests considered: • FPG • random blood glucose (RBG) • 1 hour 50 g GCT. D. The assumptions used in the model of screening strategies are given below: • A 2 hour 75 g OGTT is used as the gold standard diagnostic test (refer to Section 4. including risk factor screening. Diagnostic blood test considered: • 2 hour 75 g OGTT. Risk factors that have been considered: • • • • • age ≥ 30 years age ≥ 25 years high-risk ethnic background (ethnicity.1 contains a list of the various strategies that have been considered as screening strategies for gestational diabetes.

high-risk ethnic group. and thus sensitivity analysis was undertaken to determine to what extent test acceptance determines the cost-effectiveness conclusions of the model. The baseline values reflected the views of the GDG.Appendix D • The available data on BMI are not consistent. BMI> 27 kg/m². Table D. It was assumed initially that the risk of those with a BMI greater than 25 kg/m² is equal to that of those with a BMI greater than 27 kg/m². The model assumes that women are more likely to accept a test if they have already been identified as being at higher risk. After some initial modelling. D.1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 a List of screening strategies Risk factor – ADA criteria ADA criteria ADA criteria ADA criteria ADA criteria ADA criteria – – – – – Age ≥ 30 years Age ≥ 30 years Age ≥ 25 years Age ≥ 25 years Age ≥ 30 years Age ≥ 25 years High-risk ethnicity High-risk ethnicity High-risk ethnicity a Strategy number Screening blood test – FPG RBG GCT FPG – GCT FPG RBG GCT FPG GCT FPG GCT FPG GCT – – FPG GCT – Screening diagnostic test OGTT OGTT OGTT OGTT – OGTT – – – – OGTT OGTT OGTT OGTT OGTT OGTT OGTT OGTT OGTT OGTT OGTT Having one or more of the following risk factors: age > 25 years. but clearly considerable uncertainty surrounds the actual test acceptance. If there is a genuine difference in the subpopulation (BMI 25– 27 kg/m²). Population level data on BMI from the Office of National Statistics or the Health Survey for England is presented as overweight and obese with a BMI greater than or equal to 25 kg/m².3. especially where they are required to fast for a period beforehand. 169 .2 to D.6 lists the input parameters relating to test acceptability. family history of diabetes. this assumption may overestimate the number of cases of gestational diabetes in the at-risk population and lead to a greater number of false positive diagnoses of gestational diabetes.2 Screening strategy input parameters The parameters used to populate the model have been chosen based on the best available evidence. though this assumption could be relaxed in sensitivity analysis. Some women may find the tests inconvenient and unpleasant. either by risk factor or a previous screening test.5. the GDG expressed concern that test acceptability might be an additional important consideration. and those relating to screening are listed in Tables D. Table D. whereas the data presented in the literature129 used a BMI greater than or equal to 27 kg/m² to define some at risk of gestational diabetes based on BMI.

1 36.8 (ONS.2 (Davey and Hamblin. 1994)453 PPV (%) 10. (2002)124 Updated from Scott et al. (2002)124 Risk factor screening FPG RBG 1 hour 50 g GCT 2 hour 75 g OGTT Table D.5 High-risk ethnic group 8.5 (ONS.6 Test FPG RBG Test acceptance Initial test acceptance 0. 2001) 39.5 Gestational diabetes in 3. (2002)124 Updated from Scott et al.00 1.Diabetes in pregnancy Table D. (2004)451 Gold standard 1 hour 50 g GCT 2 hour 75 g OGTT Table D.90 GDG estimate GDG estimate GDG estimate GDG estimate 1 hour 50 g GCT 2 hour 75 g OGTT 170 .40 Test acceptance if identified as ‘at risk’ Source 0.2 Test FPG RBG Accuracy of screening and diagnostic blood tests Sensitivity 0.65 (Coustan.88 0.0 68. 2001)129 3.85 (Coustan.00 0.7 (ONS.0 (Davey and Hamblin.43 1. 1993) 452 Age ≥ 30 years Age ≥ 25 years 4.5 Risk factor Risk factors for gestational diabetes other than age % of population (Source) % of women with gestational diabetes (source) 30 (Weeks et al..58 Source GDG estimate Updated from Scott et al.90 1.7 (Davey and Hamblin.39 £10. 2001)129 28.0 Source Reichelt et al.5 (HES.97 0.7 4. 2001)129 14. 2001)129 35.39 £5. (1998)449 Ostlund and Hanson (2004)450 Seshiah et al.9 (Davey and Hamblin.80 1.78 0. 1993)452 0.0 Table D.2 (ONS.4 Risk factor Risk factors for gestational diabetes – age % of population (Source) % of women with gestational diabetes (source) PPV (%) 48.0 Specificity 0. (2002)124 Updated from Scott et al.3 Variable Cost of screening and diagnostic blood tests Cost £2 £5.50 0. 2005) a previous pregnancy Family history of diabetes BMI ≥ 27 kg/m² 10. 2005) 74. 2005) 0.48 0.61 £28.90 0. 2001) Table D.70 0.

Appendix D D. The detection rate of a risk factor screen is given by the true positive rate. but an undesirable consequence of screening may be the unnecessary inconvenience and worry associated with false positives. This detection rate is an important component of the model. There may be some desirable trade-off between detection and unnecessary testing and treatment. 171 . Its flip-side (false negatives) is also important because there may be ‘downstream’ costs associated with missed cases. The literature tends to focus on the detection rates of a particular risk factor (or more rarely combination of risk factors). TP = true positive. Figure D.  This information obviously also gives the proportion who do not meet the criteria. Therefore. To  In our gestational diabetes model.2).3 Incorporating risk factors within the model General overview In terms of the decision tree for the gestational diabetes screening/treatment model. Therefore. FP = false positive. risk factors can be thought of analogously to diagnostic tests (Figure D. given data limitations. Using Office for National Statistics (ONS) data in combination with the literature it is possible to estimate the true positive. In the economic model of screening we are also concerned with the unnecessary costs of screening which are caused by false negatives. false positive. However. true negative and false negative rates for a single risk factor screen at baseline prevalence. this is complicated by assumptions made about test acceptance.  It is not explicitly addressed in the model. the model has been developed to explore how the conclusion may vary at different disease prevalence. FN = false negative. it is much more difficult to derive these estimates for screening strategies based on combinations of risk factors.3. The screening does not lead to improved outcomes in women with gestational diabetes and the scarce resources used in screening have an opportunity cost in terms of the benefit they could have achieved if used elsewhere in the healthcare system.2 Decision tree for risk factors. as treatment costs and effects are predicated on it. the screening strategy with the highest detection rate is not necessarily the most costeffective. TN = true negative Positives from a risk factor screen or screen/diagnostic test progress to the next stage of testing or treatment. false positive. The methodological problem The data requirements for the model for any risk factor screening strategy are conceptually straightforward: • what is the disease prevalence? • what proportion of the population meets the risk criteria? • what proportion of cases is detected in the population who meet the criteria? With answers to these questions the true positive. true negative and false negative branches of the decision tree can be completed. Negatives do not progress. Prevalence varies across the country and this is potentially important in the cost-effectiveness of screening as it influences the trade-off between detection and false positives.

which would change the PPV in those of ‘at risk’ age. Next. there would be concomitant increases in the proportion with multiple risk factors. risk factor proportion would be the dependent variable. However. in this model. Prevalence = a + b1RF1 + b2RF2 + b3RF3 + … + bnRFn Such a model could be used to predict individual risk of disease. This is even true for the ADA strategy. The prevalence in the low-risk group is given by 1 − NPV.e. These estimates use a combination of the literature and ONS data and they are probably reasonably good at baseline because they are not based on a simplified model extrapolation. However. This means that the most costeffective screening strategy may be determined by the demographic characteristics of a particular population rather than prevalence per se (although the latter is a function of the former). Logically. or to conduct sensitivity analysis. Each risk factor screening strategy involves dividing the population in two – those at ‘high’ risk and those at ‘low’ risk. This would exert an upward pressure on prevalence over and above that arising from the change in a single risk factor.  ADA may be a slight exception because the paper used to derive PPV and NPV values was based on a US population with a lower prevalence than our baseline model. this linear relationship between risk factor proportion and prevalence is clearly a simplifying assumption in this case. for example. In practice. age) increases. a relatively young pregnant population may have the same gestational diabetes prevalence as an older pregnant population. This poses further methodological difficulties because of the complex and interdependent relationship between risk factors. it is possible to envisage a multiple regression equation which would predict the change in prevalence arising from a change in the proportions with different risk factor (RF) combinations. Such sensitivity analysis may indicate to what extent the simplifying assumptions drive the cost-effectiveness conclusions. it should also be noted that the software developed for modelling included an option to override the relationship between prevalence and risk factors. With sufficient individual level data. Our approach to modelling risk factor screening Owing to data limitations and methodological complexity. 172 . In a hypothetical scenario where there was just one risk factor for a disease. Recognising these simplifying assumptions as a limitation. what proportion of the high-risk group had gestational diabetes? This gives the prevalence of gestational diabetes for the highrisk group. if the younger population has a higher proportion in high-risk ethnic groups. However. The model does not capture the impact and interdependence of multiple risk factors. if known. We then assume that the PPV and NPV are independent of prevalence. where one has a higher level of risk than the other.Diabetes in pregnancy do this required that we model a relationship between changes in disease prevalence and the proportion classed at ‘high risk’. changes in gestational diabetes prevalence are likely to lead to a smaller change in risk factor proportion than that implied by the model. as clearly there is no reason why the proportion with multiple risk factors should be constant with respect to prevalence. So. The weights are the proportions in each of the groups. Similarly. it is likely that different combinations of risk factors are consistent with the same overall disease prevalence. in reality .e. our approach involved certain simplifying assumptions and the accuracy of the model may ultimately depend on whether these give a sufficiently good approximation to the real world.g. the disease prevalence is the weighted average of the respective prevalence in these two groups. As a result any model change in gestational diabetes prevalence would lead to a change in risk factor proportion. If this option were chosen. Prevalence = (proportion high risk × high-risk prevalence) + (proportion low risk × low-risk prevalence) The first step is to estimate a PPV for each risk factor screen. this would be correct. As the proportion with a risk factor (e. the user of the software would themselves select the ‘at risk’ proportion and the proportion of cases that would exist in this population. what proportion of the low-risk group did not have gestational diabetes. This can be used to reflect better local data. an NPV is calculated. i.  ‘High’ and ‘low’ risk should be interpreted as a comparison of two groups. if the low-risk group have some risk factors then their disease prevalence (1 − NPV) is also likely to change with the demographic differences associated with changing disease prevalence. i.

462 suggested 4% of gestational diabetes cases would be missed by ADA criteria. all the population would be high risk as defined by ADA and therefore this is what our model assumes for the baseline prevalence (3. we needed to model the relationship between ADA parameters and prevalence even for our baseline analysis.454 which compared selective screening (using ADA criteria) with universal screening. The PPV in conjunction with the highrisk proportion gives the detection rate.5%). using a retrospective study by Danilenko-Dixon et al. The authors estimated that only 10% would be exempt from screening in their population (of which 17.7). The gestational diabetes prevalence in the low-risk group is given by 1 − NPV (0.7 Parameter n Prevalence = 564 / 18 504 High risk = 0.9 × 18 504 Gestational diabetes cases in high risk = 564 − 17 PPV = 547/ 16 654 Low risk = 0. it is therefore possible to estimate the proportions in the high-risk and low-risk categories. The key assumption in modelling this was to assume that the PPV and NPV were independent of disease prevalence. 173 . having none of the ADA risk factors. given the study on which our calculations were based. The PPV is essentially the disease prevalence in the high-risk group. Ethnicity: Here ‘high risk’ is defined as women in a ‘high’ prevalence ethnic group and ‘low risk’ is defined as women in a ‘low’ prevalence ethnic group.28% 1. The overall prevalence can then be seen as a weighted average of the high-risk and low-risk groups. The prevalence of gestational diabetes in their population was 3% (564/18 504) (see Table D.05% 16 654 547 3.92%).850 17 99. For a given population gestational diabetes prevalence. Table D.8% were under 25 years).e. ADA criteria: ADA selective screening criteria exclude women who are: • • • • age < 25 years BMI < 27 kg/m² low-prevalence ethnic group no first-degree relative with history of diabetes. i. > 90% proportion ‘high risk’ and > 97% gestational diabetes detection might be considered ‘realistic’ .1% In this case. They found that 3% (17/564) of gestational diabetes cases were missed using ADA criteria.1 × 18 504 Gestational diabetes cases in low risk NPV = 1833/1850 Calculating PPV and NPV using ADA criteria as a risk factor screen Value 18 504 3.  However.  Another study by Williams et al.28% or more. What this modelled relationship implies is that for prevalence of 3. because the calculations are taken from a population having different disease prevalence. This would not be the case in reality for reasons outlined in the preceding section. The PPV and NPV for the ADA criteria were calculated as follows.Appendix D Below we outline in more detail the assumptions that were made for each risk factor screening strategy used in the model.

5% 68.1% 98. BMI of 27 kg/m² or more: This strategy identifies high-risk women as having a BMI of 27 kg/m² or more and low-risk women has having a BMI of less than 27 kg/m².035 22 604 8183 14 421 3.362 645 835 231 209 414 624 0. it is possible to estimate the high-risk ethnic group proportion from any given population gestational diabetes prevalence. The model suggests that at a population prevalence of 2%. At a gestational diabetes prevalence of 10% it predicts 32. Therefore.98%. As with the ADA creiteria. 174 .8.5% Source ONSa Davey and Hamblin (2001)129 ONS Calculated Calculated GDG estimate Calculated Calculated Calculated Calculated Calculated ONS data are based on proportions with BMI > 25 kg/m² (see earlier discussion).Diabetes in pregnancy The approach we used was similar to that used for the ADA criteria and is described in Table D.8 Parameter Proportion of high risk Proportion of gestational diabetes high-risk ethnic group Births Births high-risk ethnic groups Gestational diabetes prevalence Gestational diabetes births Gestational diabetes births high-risk ethnic groups PPV (15 529/54 896) NPV (583 864/590 939) Calculating PPV and NPV in using high-risk ethnicity as a risk factor screen Value 8. the high-risk ethnic proportion would be 2. these seem fairly plausible estimates but with the caveat that they are derived from a high-risk prevalence which is much higher than the literature would suggest.1% for ‘high-risk’ ethnic groups seems considerably higher than values quoted in the literature.  A prevalence of 28.5% 96.5% 22 604 15 529 28. On the face of it.9. Table D.7% 645 835 54 896 3.8% Source ONS Weeks et al. (1994)453 ONS Calculated GDG estimate Calculated Calculated Calculated Calculated Again it was assumed that PPV and NPV were independent of disease prevalence.6%.9 Parameter High-risk BMI proportion Proportion of gestational diabetes high-risk BMI Births High-risk BMI births Low-risk BMI births Gestational diabetes prevalence Gestational diabetes births High-risk BMI gestational diabetes births Low-risk BMI gestational diabetes births PPV (8183/231 209) NPV (400 203/414 624) a Calculating PPV and NPV using a BMI of 27 kg/m² or more as a risk factor screen Value 0.358 0. these provide prevalence in the high-risk and low-risk group with the overall population prevalence being a weighted average of the two. Table D. The proportion of high-risk women in this strategy at baseline was calculated as shown in Table D.

Age ≥ 30 years: The method is the same as for age ≥ 25 years.Appendix D Family history of diabetes: This strategy identifies high-risk women as having a first-degree relative with a history of diabetes and low-risk women has having no first-degree relative with a history of diabetes.10 0.035 22 604.01% 98.01% and above.11.5% 22 604 85% 19 214 4. At baseline this gives a high-risk proportion of 74.3% (source: ONS). but using an older age threshold to define the highrisk and low-risk proportion. 9.018 13 586 14.2% and low-risk proportion of 25. At baseline this gives a high-risk proportion of 48. Table D.85 × 22 604) PPV (19 214/478 860) NPV (163 585/166 975) Calculating PPV and NPV using age 25 years or older as a risk factor screen Value 645 835 478 860 3. which is again assumed not to change with disease prevalence.8% (source: ONS).11 Parameter Total births Total births ≥ 25 years Gestational diabetes prevalence Gestational diabetes births (0. The proportion of high-risk women in this strategy at baseline was calculated as shown in Table D. The proportion of high-risk women in this strategy at baseline was calculated as shown in Table D.0% Source ONS ONS GDG estimate Calculated Coustan (1993)452 Calculated Calculated Calculated It should be noted that the model assumes that all the population is in the high-risk category for prevalence values of 4.10 Calculating PPV and NPV using a first-degree relative with a history of diabetes as a risk factor screen Parameter High-risk family history proportion Proportion of gestational diabetes high-risk history Births Low-risk family history births High-risk family history births Gestational diabetes prevalence Gestational diabetes births High-risk family history gestational diabetes births Low-risk family history gestational diabetes births PPV (9018/64 584) NPV (567 666/581 252) Value 0.399 645 835 581 252 64 584 0.10. Table D.7% and low-risk proportion of 51.035 × 645 835) Proportion detected ≥ 25 years Gestational diabetes detected (0. 175 .6% Source Davey and Hamblin (2001)129 Davey and Hamblin (2001)129 ONS Calculated Calculated GDG estimate Calculated Calculated Calculated Calculated Calculated Age ≥ 25 years: This strategy identifies high-risk women as 25 years of age or older and low-risk women being 24 years of age or less.0% 97. The detection rate is then derived using a PPV.

3. Table D.12).7% and above.3 176 The basic treatment sub-tree .6% Source ONS ONS GDG estimate Calculated Coustan (1993)452 Calculated Calculated Calculated It should be noted that the model assumes that all the population is in the high-risk category for prevalence values of 4.5% 22 604 65% 14 693 4. false negatives.Diabetes in pregnancy The detection rate is then derived using a PPV.4 D.035 × 645 835) Proportion detected ≥ 30 years Gestational diabetes detected (0. D.7% 97. The screening part of the model produces an output of true positives. Figure D. false positives and true negatives and these numbers then inform the probabilities attached to given patient treatment pathways following a positive or negative diagnosis of gestational diabetes.1 Treatment Treatment decision tree The basic decision tree for treatment is depicted in Figure D.65 × 22 604) PPV (14 693/314 512) NPV (323 412/331 323) Calculating PPV and NPV using aged 30 years or older as a risk factor screen Value 645 835 314 512 3.12 Parameter Total births Total births ≥ 30 years Gestational diabetes prevalence Gestational diabetes births (0.4. which is again assumed not to change with disease prevalence (Table D.

were also included in the model. If women are deemed not to have achieved adequate control with diet. medical treatment (insulin analogue.Appendix D As far as possible. Insulin analogue • 45 minutes of instruction from a diabetes specialist nurse • daily insulin dose: 20 units • pre-filled disposable injection device • twice-daily injections (two needles per day of treatment) • a proportion of women will experience hypoglycaemia and a small proportion of these will be severe cases requiring an inpatient admission • self-monitoring of blood glucose. Diet Initial treatment aims to control blood glucose using diet. At this 10 day assessment. as this is what the effectiveness data were based upon. This part of treatment consists of: • 30 minutes of individualised dietary advice from a qualified dietitian • 30 minutes of instruction on self-monitoring of blood glucose provided by a specialist nurse (band 5/6) • self-monitoring of blood glucose. Figure D. It is assumed that women with gestational diabetes would start treatment at a gestational age of 27 weeks and that this would continue for 90 days. Glibenclamide and metformin. The basic tree structure for an oral hypoglycaemic treatment. such as glibenclamide.4.4 Glibenclamide treatment sub-tree 177 . with self-monitoring of blood glucose reduced to twice daily. glibenclamide. The treatment protocol used in the model is outlined below. The effectiveness of metformin is currently being investigated as part of the ongoing MIG trial and is therefore a potential treatment option. and assumes one lancet and one test strip per reading) • 5 minutes of assessment of control after 10 days on diet by a specialist nurse. they remain on dietary control until the end of their pregnancy. treatment was modelled according to the ACHOIS protocol. women with gestational diabetes are judged to have achieved adequate control with diet or not. Glibenclamide • daily dose: 15 mg. would be as illustrated in Figure D. two times daily. metformin) is then initiated.5 g. An RCT of glyburide (glibenclamide) versus insulin for gestational diabetes showed no statistically significant differences in outcomes. two alternative oral hypoglycaemic treatments to analogue insulin. If they have achieved adequate control. four times daily (costing of self-monitoring of blood glucose includes one monitor. Metformin • daily dose: 1.

Diabetes in pregnancy D. Table D. a weighted cost and QALY was calculated for a serious perinatal complication based on the QALY and costs associated with each of the individual components.2 Outcomes and downstream costs The model uses the following outcomes presented in the ACHOIS study to estimate the incremental QALY gain associated with screening. that the proportion of serious perinatal complications accounted for by individual components did not differ according to whether the women were treated for gestational diabetes or not.14 to D.14 Treatment timeframe Variable Value (days) Source Notes Treatment 90 duration Exclusive 10 diet 4 × daily SMBG 10 Diabetes in The diabetes in pregnancy GDG consensus was that treatment would usually pregnancy GDG commence between 26 and 28 weeks of gestation. 178 .03 0.09 D. Table D. based on the lack of statistical significance for any difference.3 Treatment model parameters The baseline parameter values for all model treatment inputs are shown in Tables D. The trial data allow this to be easily done for deterministic sensitivity analysis.20. In order to reflect the individual components of the composite measure.72 0. ACHOIS153 The ACHOIS protocol suggested that SMBG be done 4 × daily until glucose levels had been in the recommended range for 2 weeks. it was assumed. Therefore.09 0.4. Furthermore. In order to calculate the weights.13 ACHOIS trial outcome data for serious perinatal complications combined across control and intervention groups Outcome All serious perinatal complications Stillbirth Neonatal death Shoulder dystocia Bone fracture Nerve palsy Total 32 3 2 23 1 3 Weight 1.00 0. Taking the midpoint of 27 weeks. diagnosis and treatment of gestational diabetes: • stillbirth • neonatal death • maternal health state utility. Costs are assigned to these outcomes and included in the evaluation of incremental costs: • • • • • • • neonatal death shoulder dystocia bone fracture admission to neonatal unit jaundice requiring phototherapy induction of labour caesarean section.06 0. the following outcomes from ACHOIS are assumed to have downstream cost implications. the data on individual events were pooled across both arms of the trial in order to estimate the weighting for individual components (Table D. We used the outcome data of ACHOIS for ‘serious perinatal complications’ as the measure of the effectiveness in the model. Diabetes in The diabetes in pregnancy GDG suggested that diet alone would be given pregnancy GDG 7–14 days to achieve adequate control.13). with the different event rates giving well-defined relative risks.4. 90 days is a reasonable approximation of the typical time to term. SMBG = self-monitoring of blood glucose.

assessment/review Insulin instruction Risk factor screening questions SMBG = self-monitoring of blood glucose. £1.50.85. £15.16 Variable Self-monitoring of blood glucose and treatment costs Cost £7. all similarly priced.57 for a pack of 100 needles This is based on a dose of 20 units per day.10 £403 Source BNF 52 (2006) 457 Notes Many makes. 179 .46.79 £0.248 Cost for neonate with one minor diagnosis (HRG N03) Cost for neonate with one minor diagnosis (HRG N03) Cost for neonate with one minor diagnosis (HRG N03) Cost for neonate with one minor diagnosis (HRG N03) Incremental cost over and above that of a normal vaginal birth Incremental cost over and above that of a normal vaginal birth £2.Appendix D Table D.55 for a pack of 50 was the cheapest found from a small sample.09 each £0.5 g daily.568 NHS Tariff 2006 NHS Reference Costs 2004 Shoulder dystocia £629 Bone fracture Nerve palsy Phototherapy £629 £629 £629 NHS Tariff 2006 NHS Tariff 2006 NHS Tariff 2006 NHS Tariff 2006 Emergency £1.676 NHS Reference Costs 2004 Induction of labour £20 Neonatal death Davies and Drummond Updated to 2006 prices using Retail Price Index published by ONS.17 Variable Admission to neonatal unit Downstream outcome costs Cost Source Notes Assume 2 days of neonatal intensive care at £838 per day. For remaining 75% assume 2 days of neonatal intensive care (£838 × 2) and neonate with one major diagnosis which has an NHS Tariff of £1. Based on 1.31 each £0. Blood glucose monitor Test strips Lancets Needles Insulin analogue (Humalog®) Glibenclamide Metformin Treatment of severe hypoglycaemia BNF 52 (2006)457 BNF 52 (2006)457 BNF 52 (2006)457 BNF 52 (2006)457 BNF 52 (2006)457 BNF 52 (2006)457 Curtis and Netten (2006)456 NHS Reference Costs 2005–06 Table D.572 = £3.16 £0.572.676 + £1.03 each £0. Average cost per patient journey for paramedic ambulance: £323. A 5 mg 28 tablet pack costs £1.15 Variable Dietary advice Cost of healthcare professionals’ time Time (minutes) 30 30 5 45 2 Cost per hour £28 £63 £63 £63 £63 Source Curtis and Netten (2006)456 Curtis and Netten (2006)456 GDG estimate Curtis and Netten (2006)456 GDG estimate Curtis and Netten (2006)456 GDG estimate Curtis and Netten (2006)456 GDG estimate Notes Unit costs of a dietitian for an hour of client contact Unit cost of a nurse specialist (community) for an hour of client contact Unit cost of a nurse specialist (community) for an hour of client contact Unit cost of a nurse specialist (community) for an hour of client contact Unit cost of a nurse specialist (community) for an hour of client contact SMBG instruction Control with diet. (1991)458 and (1993)459 From NHS Reference Costs 2004 finished consultant episode (FCE) data assume that 25% of neonatal deaths are < 2 days (n = 974). Table D.39 per day £0.205 NHS Reference Costs caesarean section 2004 Elective caesarean £822 section NHS Reference costs 2004 HRG = Health Resource Group. Accident and emergency department admission with low-cost investigation: £80. £8. A 500 mg 84 tablet pack costs £2. A pre-filled disposable pen has 1500 units and costs £29. Based on 15 mg daily. NHS Reference Costs for this is £527. £1.

The origin represents the no screening/no treatment option and all costs and QALYs are measured relative to this.286 0.79 Maternal QALY – no treatment 0.96 0.5 Baseline results The baseline results from the modelling exercise are given based on a population of 10 000 pregnant women and assume a baseline prevalence of gestational diabetes of 3.21 and are plotted on a cost-effectiveness plane in Figure D.20 Variable QALYs QALY 25 Source Notes This is the approximate lifetime QALYs from 75 years lived in perfect health with QALYS discounted at 3. Assumed the same as for glibenclamide.044 0.20 0.5% per annum.18 Variable Treatment pathway probabilities Value 0.5%. Averted death (stillbirth/neonatal) Maternal QALY – treatment (during pregnancy) 0. (2000) 172 74 Notes – GDG member suggested that data from Southampton (their local practice) indicate a higher failure rate (23%). (2000)74 – Langer et al. ACHOIS153 It is assumed that this QALY gain covers the entire 3 months postpartum period.Diabetes in pregnancy Table D.19 Variable ACHOIS outcome probabilities Treatment value 0.014 0.5.72 Maternal QALY – no treatment 0. – Assumed the same as for glibenclamide.116 0.374 0. The total cost and QALYs generated for each strategy under the baseline assumptions are presented in Table D.20 0.613 0.087 No treatment value 0.96 Source Persson et al.70 (during pregnancy) Maternal QALY – treatment (3 months postpartum) 0. ACHOIS153 It is assumed that this QALY gain covers the entire 3 months postpartum period.02 – Langer et al.78 (3 months postpartum) D.05 requiring hospitalisation Table D.706 0. 180 . – Assumed the same as for insulin.02 0.86 0.197 0.092 Source ACHOIS153 ACHOIS153 ACHOIS153 ACHOIS153 ACHOIS153 ACHOIS153 Serious perinatal complications Admission to neonatal unit Induction of labour Elective caesarean section Emergency caesarean section Jaundice (phototherapy) Table D. – Control with diet Control with glibenclamide Control with metformin Hypoglycaemia on insulin therapy Hypoglycaemia on insulin analogue Hypoglycaemia on glibenclamide Hypoglycaemia on metformin 0. (2000)74 – GDG estimate Severe hypoglycaemia 0.142 0. ACHOIS153 It is assumed that this QALY gain persists throughout treatment. ACHOIS153 It is assumed that this QALY gain persists throughout treatment.158 0. (1985) Langer et al.

01 29.940 £191.56 26.929 £119. From this.70 19. As can be seen from Figure D.45 25.40 25. Once these strategies are excluded the remainder are again ranked in order of effectiveness.85 39.26 33. Such strategies can unambiguously be excluded (they are said to be (strictly) dominated).747 £367.21 11 1 8 9 3 13 14 19 20 12 10 15 17 16 21 2 4 5 18 7 6 a Total QALYs and cost for each screening strategy QALYs 16.46 20.763 £1.009 Screening strategya Ranked in order of effectiveness (from fewest to most QALYs).55 21.Appendix D Table D.63 17.529 £77.731 £99.5 The cost-effectiveness plane for the baseline analysis 181 .791 £838.37 33.66 27.932 £489. Moving down the list.670 £203.580 £286. the Figure D.758 £259. it is possible to calculate the incremental costs and incremental QALYs of selecting a given strategy relative to the next best strategy.33 Cost £146.902 £269.769 £345.188 £212.94 31.465 £89.96 24.39 20.341 £198.70 18.56 21.5. a number of strategies are more expensive than the two circled and yet offer a lower QALY gain.816 £304.419 £126.48 18.48 18.561 £160.753 £145.172.43 34.

Thus it is possible that strategy 6 could reasonably be argued to be cost-effective. 182 . In the analyses presented below we primarily used a series of one-way and multi-way sensitivity analyses to explore what happened when the value of one or more parameters is changed. Tables D. for example if it is believed some salient piece of information falls outside the model such as the identification of women at higher risk of developing type 2 diabetes in the future.341 Incremental QALYs 27.01 39.6 Sensitivity analysis All decision analysis models are subject to uncertainty460 and there are two common approaches to dealing with this uncertainty – making use of a reference case (that is. family history of diabetes or from a high-risk ethnic background).668 ICER £3. If the decision maker was willing to buy QALYs at the cost implied by the ICER of the less effective strategy. The ICER for strategy 21 is calculated relative to no screening/treatment and the ICER for strategy 6 is calculated relative to strategy 21. and thus the implications for our baseline results.677 £21.and low-risk populations.27 give the sensitivity analysis ICER for strategies which have not been excluded on the grounds of strict or extended dominance. This model takes as its reference case the standards for conducting economic evaluations included in the 2007 version of the NICE guidelines manual.738 QALYs 27. It is then possible to exclude certain further strategies on the grounds of ‘extended dominance’.341 £267.738 when compared with strategy 21.22 Strategy 21 6 ICER for non-dominated strategies Cost £99.677. D. The cost-effective option is the most effective strategy which falls within the willingness to pay threshold set by the decision maker. This allows us to see what happens to the model results when these values are changed. they would logically be willing to buy (and prefer) strategies where additional QALYs could be obtained at a lower cost.31 The baseline analysis suggests that a strategy of offering women from a high-risk ethnic background a diagnostic test (strategy 21) would be cost-effective when compared with not offering a screening.23 to D. Table D.01 Incremental cost £99. sensitivity analysis.Diabetes in pregnancy ICER is derived which effectively shows the cost of ‘buying’ QALYs.000 and may be considered cost-effective under certain circumstances.33 £367. The strategy of offering a diagnostic test to those women who are deemed to be at increased risk according to the ADA criteria (strategy 6) has an ICER of £21. Extended dominance occurs when a strategy has a higher ICER than a more effective strategy. BMI greater than 27 kg/m². Although it is higher than the £20.22 all but two of the strategies are excluded on these dominance grounds. In Table D.009 12. based on the presence of one or more of the risk factors highlighted in the ADA criteria (age over 25 years. it is comfortably under the maximum willingness to pay per QALY of £30.23 The methods and assumptions used in the model are highlighted above in detail and were tested using a second method of examining uncertainty.000 per QALY threshold suggested by NICE. The analyses that follow explore the uncertainty in a number of key areas. including: • the reliability of the trial data on which the likelihood of an event occurring was based • the prevalence of gestational diabetes in the population • the proportion of women that would undergo a screening or diagnostic blood test if it were offered as a first-line test or based on identification of a potentially high-risk population • treatment options • the efficacy of using risk factors to define high. with an ICER of £3. a standard of good practice) and sensitivity analysis.

80 23.66 3.439 183 .386 £100.392 £21.347 £3.045 £62.03 Incremental cost £48.738 Table D.694 £149.65 11.852 Table D.031 £100.915 £10.131 £251.379 £251.854 £366.95 15.12 24.97 44.97 Cost £51.64 12.69 15.227 £100.209 £9.042 £16.23 diabetes Strategy Four deaths 21 6 Three deaths 21 6 Two deaths 21 6 One death 21 6 No deaths 21 6 Effect on ICER of varying the number of perinatal deaths attributable to gestational QALYs 21.93 49.490 £367.43 Cost £130.94 8.473 £368.737 £100.856 £100.26 Strategy 21 11 1 Gestational diabetes prevalence of 2% and 100% test acceptance QALY 10.079 £19.69 4.085 £19.049 £172.87 5.26 30.113 Incremental QALY 10.115 £4.43 4.385 £163.069 £369.634 £31.856 £51.85 Incremental cost £130.657 £101.183 £249.507 £44.34 Cost £99.80 7.41 12.41 3.633 £6.727 £76.434 £336.62 56.816 £411.825 £401.97 10.634 £161.069 £268.583 £177.738 £21.87 Cost £48.69 15.61 0.56 5.27 Strategy 19 21 1 Gestational diabetes prevalence of 5% and 100% test acceptance QALY 41.71 1.58 62.46 21.184 £101.385 £55.94 2.01 10.85 Incremental cost £51.913 £99.18 Cost £113.205 Incremental QALYs 33.336 £37.680 £27.473 £268.56 11.136 £268.46 10.24 Strategy 21 2 6 Gestational diabetes prevalence of 2% QALYs 9.Appendix D Table D.456 ICER £4.694 £36.20 10.167 £100.272 Table D.25 Strategy 19 21 6 18 Gestational diabetes prevalence of 5% QALYs 33.56 Incremental cost £113.536 ICER £5.679 ICER £4.766 ICER £3.192 £15.287 £368.18 56.93 7.100 £100.005 Table D.118 Incremental QALYs 9.205 £401.84 16.379 ICER £3.26 9.385 £112.583 Incremental QALY 41.490 £267.287 £268.136 £368.65 1.525 Incremental QALYs Incremental cost 21.73 £99.61 2.

However. especially as with default assumptions there is an inverse relationship between test accuracy and test acceptance.6. Tables D. This difference was not statistically significant.6. strategy 6 would remain the preferred option up to a willingness to pay threshold of £20. D. This is because of the non-negligible weight given to this outcome as a proportion of all serious perinatal complications and the significant gain in QALYs to be made by preventing a perinatal death.27 show how the results varied when it was assumed that all women were tested in populations with a relatively low and relatively high disease prevalence.26 and D.000 per QALY. A threshold analysis. its advantages in terms of detection rate are negated if it is assumed that the test has a low level of acceptance.1 Outcomes The outcome that had the greatest influence on the model results was the number of perinatal deaths (stillbirths and neonatal deaths). However. 2. The number of deaths in the control group was similar to the number of perinatal deaths that would be expected in the general population according to ONS data on perinatal mortality (in 2005 there were 5. showed that. the ICERs when no deaths are assumed are sufficiently large to suggest that the potential QALY gain from preventing some of these events would not be adequate for these strategies to be cost-effective. but it should be remembered that the simplified model relationship between risk factor proportions and gestational diabetes prevalence has a bearing on these results.24 and D. diagnosis and treatment that could be considered cost-effective.6. However. with all other model parameters at their baseline values. 184 . The results suggest that varying the prevalence over a range of 3 percentage points has little impact on the cost-effectiveness conclusions of the model. and may therefore underestimate the cost-effectiveness of screening. respectively. diagnosis and treatment strategies are highly influenced in the model by this one particular adverse outcome. As the number of perinatal deaths decreases. Tables D. Table D. The authors of the ACHOIS study highlight that at least one death in the control group was unrelated to gestational diabetes. if no perinatal deaths are attributed to gestational diabetes. then there is no strategy for screening. test acceptance rates are potentially an important source of uncertainty within the model. When only four deaths in the trial group are attributed to gestational diabetes. D.4 stillbirths. The results show that a universal screening strategy using the gold standard diagnostic test becomes more cost-effective as disease prevalence increases. This result demonstrates that the model is highly sensitive to the potential QALYs gained by preventing even a single perinatal death.3 Test acceptance As noted earlier. and the cost-effectiveness of screening.23 shows the results of the models when the number of perinatal deaths in each group was assumed to be different to that reported in the ACHOIS trial. What is clear from this analysis is that the potential benefits to the NHS with respect to QALYs gained from intervention are likely to be felt in the form of preventing perinatal deaths.25 show how the results of the model varied for different prevalences of gestational diabetes. such as shoulder dystocia or nerve palsy. the ICERs of both strategies 21 and 6 become less favourable and this continues until only one perinatal death is attributed to gestational diabetes. the cost-effectiveness of the various strategies changes.2 Gestational diabetes prevalence The prevalence of a disease can often be a very important determinant of the cost-effectiveness of screening.6 early neonatal deaths and 3. In the ACHOIS trial there were five perinatal deaths recorded in those who received no treatment (n = 524) while in the treatment arm there were none (n = 506). diagnosis and treatment strategy that would be considered cost-effective – in this case strategy 21. even if test acceptance for FPG/OGTT in women identified as ‘at risk’ fell from 90% at baseline to 52%. The model also potentially underestimates the QALYs to be gained by preventing other adverse outcomes.4 late neonatal deaths per 1000 total births in England and Wales). This is because of its advantages over other test options in terms of its detection rate. Even when there is only a single death assumed. there is still a screening.Diabetes in pregnancy D.

and because all the incremental analysis is undertaken with treatment cost as a given.29 15. has been performed.109 £81.73 Cost £66. This would be a considerable inconvenience to a large number of women. as well as putting a strain on local services.000 and could in each case be considered cost-effective on its own.28 shows that the choice of treatment option in the model made little difference to the ICERs for the screening strategies.935 £12. it may be the case that where single risk factors are cost-effective on their own. the GDG expressed concerns over the number of women that would have to undergo a OGTT if strategy 6 were adopted.5 Single risk factors The baseline analysis suggested that strategy 6 was a borderline cost-effective strategy using a willingness to pay threshold of £20.29 ICER for single risk factor strategies followed by a diagnostic test when compared with a strategy of no screening or treatment Strategy Ethnicity BMI Family history QALY 9. Table D. followed by an OGTT.6.29. As a result it was decided that the use of screening based on risk factors other than age should be considered. The above analysis established that screening. A large proportion of women tested would be tested based on age criteria alone – under the baseline assumptions as many as 90% might be offered the diagnostic test. However. Below we consider the cost-effectiveness of the various treatment options for gestational diabetes.647 £21. has an ICER that is below the threshold of £20. Table D.226 £80. then any combination of these is also likely to be cost-effective. with each risk factor plus OGTT combination compared with a strategy of no screening or treatment.000 per QALY. The results are presented in Table D. Therefore an analysis of the cost-effectiveness of each single risk factor. followed by a diagnostic test.Appendix D D. a lower treatment cost will reduce the cost of each strategy but may have relatively little impact on the incremental costs.55 6.6. This is because treatment represents a relatively small proportion of the total costs. diagnosis and treatment of gestational diabetes is generally cost-effective in some populations.738 £21. glibenclamide and metformin). However. The PPVs and NPVs of different risk factor combinations are not accurately known which means that the relative cost-effectiveness of different combinations of any of the single risk factors can not be calculated.915 ICER £6.4 Treatment option The model also allowed the ICERs for different strategies to be calculated for different treatment options (analogue insulin.736 £5. Table D.208 Any strategy where a single risk factor from the ADA criteria other than age is applied alone. only a small minority of whom would ultimately benefit from the testing process.28 Treatment Analogue insulin Glibenclamide Metformin ICER for strategy 6 for different treatment options in the baseline model ICER £21. 185 .642 D. For example.

The costs for a woman taking insulin analogue include the time of a diabetes specialist nurse in providing instruction on how to administer the drug.9 g. Included in the costs for insulin were drug costs. the cheapest option is unambiguously cost-effective as it dominates the alternatives. The cost of glyburide was based on the average wholesale cost of a milligram of drug multiplied by the weekly dose expected to be necessary for glycaemic control. Metformin was additionally added into this analysis as the ongoing MIG study is assessing its use in women with gestational diabetes and it could potentially be an important treatment option in the UK.2 Method The basic structure of the cost analysis is shown in Figure D. it was assumed that 4% of patients would not achieve control with glyburide and would have to switch to insulin. 186 . As described in the screening. women with gestational diabetes would start with dietary treatment. costs of the consumables needed to administer the insulin and the cost of instructing women with diabetes on how to draw up the insulin and inject themselves. which was assumed to be more common in insulin-treated gestational diabetes.7. Humalog® Mix50) and to be on a daily dose of 20 units administered in twice-daily injections. It is assumed that a diagnosis of gestational diabetes would be made at a gestational age of 27 weeks. D. In addition. targeted at the guideline question on what is cost-effective treatment for gestational diabetes. are not included in the analysis. no evidence of a difference is not the same as evidence of no difference. Therefore. Costs which are common to all treatments.5 g. the cost of metformin is based on a daily dose of 1. pharmacological therapy would be started and this is the starting point for the cost comparison.461 This paper described a cost model to compare the costs of an oral hypoglycaemic. but the P values in this study were particularly large and the inference of no difference does not arise as a result of some outcomes being just the wrong side of an arbitrary 5% cut-off point for statistical significance.7. such as those associated with self-monitoring of blood glucose. diagnosis and treatment model. Of course.g. glyburide produced an average cost saving of $166 per woman. glyburide (glibenclamide). In women who do not achieve adequate glycaemic control after 10 days. A similar cost model was developed to compare the cost of insulin analogue (lispro) with that of two oral hypoglycaemics (glibenclamide and metformin) in a UK context. Insulin analogue was used in this cost comparison rather than insulin. Implicit in this is an assumption that outcomes with an insulin analogue would be equivalent to those with insulin. being cheaper and equally effective. Therefore. Finally.Diabetes in pregnancy D.74 Their model. and in the absence of any conflicting evidence. the model also included the downstream costs of hypoglycaemia. that such a cost minimisation analysis might be justifiable to determine the cost-effectiveness of various gestational diabetes treatments. In such a scenario. excluded resource items that were identical in both treatments. identified a single paper for inclusion. D. as this is what would be offered to women with gestational diabetes in the UK.24 per week in conjunction with a daily dose of 18. the model also incorporated a cost for glyburide treatment failure. Women with gestational diabetes are assumed to use a pre-filled disposable injection pen (e.7 Cost analysis of different treatment options for gestational diabetes A systematic review of the literature. with those of insulin for the treatment of gestational diabetes. based in a US setting. The paper justifies what is essentially a cost minimisation approach on the basis that glyburide and insulin confer similar glycaemic control. A threshold analysis suggested that insulin was only less costly than glyburide at the highest wholesale cost of $18. they require two needles per day for their injection pen. It is on this basis. A randomised study74 failed to find significant differences in outcomes (maternal and neonatal) between glyburide and insulin treatment in women with gestational diabetes. Similarly. which is considerably higher than what is believed to be necessary to achieve good glycaemic control. Women switching to insulin also incurred the educational costs associated with insulin treatment.6.1 Introduction A cost minimisation analysis can be considered to be a special case of cost-effectiveness analysis when the interventions being compared are equally efficacious. The cost of glibenclamide is the drug cost based on a daily dose of 15 mg. In the baseline analysis. The authors reported that most sensitivity analyses did not alter the direction of this finding.

Appendix D Figure D. D.7. It is additionally assumed at baseline that 5% of hypoglycaemic events will be ‘severe’ and it is these for which there will typically be an NHS resource implication. but following the Langer study74 the model addresses a possible differential in the hypoglycaemia risk between the different treatments. These show the oral hypoglycaemics to be considerably cheaper than analogue insulin.6 Gestational diabetes treatment cost model In addition to the cost of treatment it is important also to consider downstream costs. The cost of a ‘severe’ hypoglycaemic event is assumed to be the cost of a paramedic ambulance journey and an admission to an accident and emergency department. threshold analyses were also undertaken which showed that. metformin is the cheapest and.7 shows how the incremental cost of insulin analogue varies with different assumptions about the proportion of women with gestational diabetes who achieve adequate glycaemic 187 .33 lists the cost per patient of each of the three treatment options. holding all other factors constant.4 Sensitivity analysis A number of sensitivity analyses were undertaken to determine how robust the conclusion of the baseline result was to changes in model parameters where some uncertainty exists as to their ‘true’ value. metformin remained cheapest as long as control on metformin was at least 90.3% (with control on glibenclamide 77%). For ease of exposition. Figure D.30 to D. most sensitivity analyses focus on a comparison of glibenclamide and insulin analogue on the basis that. The complete list of model parameters is given in Tables D. the most costeffective treatment.32.3% (with control on glibenclamide 96%) or control on metformin was at least 72. Overall outcomes are assumed not to differ.7. Of the oral hypoglycamics. D.3 Results Table D. However. with the assumption of equal clinical effectiveness. these are assumed to be identical treatments in terms of both outcomes and downstream costs. apart from a small difference in costs.

Assumed to be the same as Langer found for insulin.09 each.77 for this parameter in his clinical practice. (2000) 74 Notes A GDG member reported 0.00 oral hypoglycaemia failure Treatment of severe hypoglycaemia £403 Curtis and Netten Average cost per patient journey for paramedic ambulance: £323.Diabetes in pregnancy Table D. GDG estimate This is based on a dose of 20 units per day. A 5 mg 28 tablet pack costs £1. (2000)74 Langer et al. (2000)74 GDG estimate Hypoglycaemia on metformin 0.05 that is ‘severe’ Table D. Costs 2005–06 Table D.57 per BNF 52 (2006)457 day £0.33 Cost per woman with gestational diabetes Treatment Insulin analogue Glibenclamide Metformin Average cost per woman with gestational diabetes £96.10 BNF 52 (2006)457 BNF 52 (2006)457 GDG estimate Glibenclamide Metformin Switching cost of £0.92 £16. – Langer et al.46. A 500 mg 84 tablet pack costs £2. – Assumed identical to glibenclamide. Based on 15 mg daily. It is further assumed that injections are twice daily.31 Variable Costs Cost £47.85.32 Probabilities Variable Control with glibenclamide Control with metformin Hypoglycaemia on insulin analogue Hypoglycaemia on glibenclamide Probability 0.96 0. (2000) GDG estimate Langer et al. requiring two needles at £0. It is assumed there is no additional cost over and above those incurred by all women with gestational diabetes starting insulin analogue treatment.20 0.50.5 g daily. Women with gestational diabetes would be given approximately 10 days to achieve control with diet and 80 days is a reasonable approximation of the typical time to term at the commencement of pharmacological treatment. Insulin instruction Insulin analogue £0.02 Source Langer et al. (2000)74 74 Langer et al. A pre-filled disposable pen has 1500 units and costs £29. Assumed identical to glibenclamide.30 Treatment timeframe (days) Variable Treatment duration Value (days) 80 Source Diabetes in pregnancy GDG Notes It is assumed a gestational diabetes diagnosis would be made at 27 weeks of gestation.25 Source Notes Curtis and Netten This is based on an instruction time of 45 minutes with instruction (2006)456 provided by a specialist nurse.02 Proportion of hypoglycaemia 0.32 £11.16 £0.96 0. Oral hypoglycaemic 14 trial period ACHOIS153 Table D. Based on 1.68 188 . (2006)456 Admission to an accident and emergency department with lowNHS Reference cost investigation: £80.

Figure D. Similarly.8 Incremental cost of insulin analogue as hypoglycaemia risk of insulin analogue varies 189 . insulin analogue continues to be the more costly option even if only 40% of women achieve adequate glycaemic control with glibenclamide.Appendix D control with glibenclamide.9 shows that the costs of treating hypoglycaemia are not an important determinant of the additional costs of insulin analogue. Figure D. Although the differential in cost declines with reduced glibenclamide clinical effectiveness.7 Incremental cost of insulin analogue as control on glibenclamide varies Figure D.8 shows that the cost analysis is not sensitive to the risk of hypoglycaemia in women taking glibenclamide. Figure D.

if it too was shown to be as effective as insulin analogue then it would be the most cost-effective treatment of all. One caveat to these findings is the assumption that there is no cost to the NHS in switching from an oral hypoglycaemic agent to insulin analogue.7. If there were a ‘switching cost’.Diabetes in pregnancy Figure D. then we can say that the results presented here suggest that glibenclamide is a more cost-effective treatment for gestational diabetes than insulin analogue. given that the treatments have different resource implications for the NHS. if it is argued on the basis of this study that glibenclamide is equally effective as insulin analogue and would have achieved similar outcomes to those observed with diet and insulin treatment in ACHOIS. However. as the sensitivity analysis shows.5 Discussion Using the data from ACHOIS.9 Incremental cost of insulin analogue as cost of treating severe hypoglycaemia varies D.74 However. The lack of differences between the infants born to mothers in the two treatment groups corroborated the results in the mothers’. other than those ordinarily incurred for women with gestational diabetes taking insulin analogue. 190 . then the costeffectiveness of the oral hypoglycaemic agents would be less than that implied here. As yet. there is no evidence to justify a cost minimisation approach with metformin. However. Therefore. it does not follow that all treatments are equally cost-effective. Sensitivity analysis suggested that this conclusion was robust when model parameters were changed in a one-way fashion. the degree of glycaemic control and the perinatal outcomes were essentially the same for those treated with glyburide (glibenclamide) and those treated with insulin. this guideline has demonstrated that screening. diagnosis and treatment of gestational diabetes is cost-effective and that this finding is not contingent on the type of pharmacological treatment used (insulin analogue or oral hypoglycaemic agent). glibenclamide continues to be cost-saving compared with insulin analogue even with a much smaller proportion achieving adequate control. One study74 suggested that ‘among women with gestational diabetes. The diabetes in pregnancy GDG has suggested that the proportion of women with gestational diabetes achieving control with glibenclamide may be lower in clinical practice than that observed by Langer et al.

A decision tree model was developed for the guideline in Microsoft Excel® to assess the costeffectiveness of second-trimester screening for congenital cardiac malformations in pregnant women with diabetes. but using a four chamber plus outflow tracts view may allow the detection of some abnormalities.415 Nevertheless. there are concerns about the generalisability of costeffectiveness results from healthcare settings outside the UK and it was. the study also noted that universal fetal echocardiogram yielded the highest detection rate of cardiac anomalies. selective fetal echocardiography after abnormal detailed anatomical survey (a four chamber view of the heart plus outflow tracts was included as part of the detailed anatomical survey). the scenarios presented did not reflect current UK practice or viable alternatives.413 In the study. However. which are not usually visible with a four chamber view. therefore. and universal fetal echocardiography. and the healthcare costs of a major cardiac malformation over and above those needed for a healthy child. and healthy neonate 1. Effectiveness was measured using QALYs derived by assigning different utilities to different outcomes (major cardiac defect 0.1% prevalence of major cardiac malformations.1 Introduction A review of the health economics literature identified a single study addressing the costeffectiveness of screening for congenital cardiac malformations in pregnant women with diabetes. fetal echocardiography for high HbA1c only. a decision-analytic model was used to compare the cost-effectiveness of four screening strategies for congenital cardiac malformations in a healthcare setting in the USA. The strategies were: no screening. terminations of pregnancy. In addition. the study reported that the model results were robust when considering parameter uncertainty. fetal death 0. such as TGA and tetralogy of Fallot. The sensitivities and specificities for each strategy were derived from a literature search and costs included tests. decided to develop a model for this guideline to compare the cost-effectiveness of two screening strategies in the UK context. there is some evidence suggesting that an antenatal diagnosis of TGA may reduce mortality. Screening for these malformations in pregnant women with diabetes is likely to be relatively more cost-effective than in pregnant women without diabetes because the prevalence of these anomalies is much higher in women with diabetes. There are difficulties in considering the cost-effectiveness of screening using termination as a positive outcome and the evidence that screening produces a survival advantage is limited. However.0) and then multiplying by the life expectancy with each outcome. Costs and outcomes were modelled for a hypothetical cohort of 40 000 pregnant women with diabetes and with a 2. As a result of a probabilistic sensitivity analysis. This is important for the health economic analysis because TGA is an anomaly that would not normally be identifiable with a four chamber view. The study reported that selective fetal echocardiogram after abnormal detailed anatomical survey dominated all other strategies with baseline assumptions.Appendix E Cost-effectiveness of screening for congenital cardiac malformations E.5.07. Current UK practice is to screen pregnant women using a four chamber ultrasound scan at 20 weeks of gestation. The utility weights were discounted at a rate of 3% per year. but it would be with the addition of 191 .414 There are two principal reasons why it may be beneficial to screen for congenital cardiac malformations: • it allows women to consider termination of pregnancy • improved outcomes for women and/or babies.

The basic decision tree structure is illustrated in Figure E.2% is for prevalence at birth. outpatient visits. TGA = transposition of the great arteries. E. (2003)414 Wren et al.2 Model parameters The parameter values used in the baseline model are shown in Tables E. If the woman chooses to continue the pregnancy then she will either give birth to a live baby or suffer a pregnancy loss. tests Four chamber plus outflow £46 tracts view ultrasound scan Fetal echocardiography Termination of pregnancy Birth £62 £492 £3. Table E. other maternity events. (2004)416 ICER = incremental cost-effectiveness ratio. 0.a Prevalence of cardiac malformations 0.a Mean value for an echocardiogram Cost of a surgical termination A weighted average including birth.Diabetes in pregnancy a view of the outflow tracts and. Wren et al.0405 Ritchie et al.000 NHS Referemce Costs 2005–06 NHS Tariff 2006/07 NHS Reference Costs 2003.2 Costs Characteristic Four chamber view ultrasound scan Cost £34 Source NHS Reference Costs 2005–06 GDG estimate Notes Mean value for a maternity ultrasound Based on estimate that appointment slots would be 20 minutes compared with 15 minutes for a four chamber view. (2004)416 Derived from survival probability from second trimester to birth.144 that are TGA in women with diabetes Pregnancy loss after 20 weeks of gestation (no cardiac malformations present) Pregnancy loss after 20 weeks of gestation (cardiac malformations present) 0. but this is not important if the four chamber plus outflow tracts view is shown to be cost-effective. Women with a positive result on either scan will then be referred for fetal echocardiography to confirm diagnosis and guide subsequent treatment. A proportion of babies born with cardiac malformations will have TGA and they may survive or die. the model particularly focuses on the cost-effectiveness of antenatal diagnosis of TGA. 192 .0032 at 20 weeks of gestation Proportion of cardiac malformations 0. If the diagnosis is confirmed then the woman has the option to terminate or continue the pregnancy.4. Table E. This is likely to represent a small bias in the model against the four chamber plus outflow tracts view. NHS General Medical Services Revised Fees and Allowances 2003–04 a The cost of the four chamber plus outflow tracts view does not take into account the fact that the number of equivocal scans is likely to increase. neonatal care.1 to E.0115 Ritchie et al. GP fees. Derived from survival probability from second trimester to birth. therefore. At 20 weeks of gestation women receive either a four chamber view ultrasound scan or four chamber plus outflow tracts view ultrasound scan. (2003)414 Wren’s value of 3.1.1 Population characteristics Characteristic Population Value 1000 Source Notes Prevalence data are often given as rates per 1000 head of population and the ICER from the model is not affected by population size. a The prevalence of cardiac malformations at 20 weeks of gestation may be slightly higher than at birth if terminations and fetal death are higher in affected than non-affected pregnancies.

Appendix E 193 Figure E.1 Four chamber plus outflow tracts view versus four chamber view decision tree for women with diabetes. TGA = transposition of the great arteries .

(2006)417 In 58 cases of congenital cardiac defects. Bonnet 1988–97. 0.pdf) 275 Notes Four chamber view specificity 1. (1997)275 Smith et al.00 view specificity TGA proportion of malformations 0. Table E.0 TGA mortality detected postnatally QALY weight successful TGA treatment Annual discount rate Results reported in presentation by Wellesley et al. five were TGA.419 Kumar 1988–96420 0. the higher sensitivity of the four chamber plus outflow tracts view results in 1.25 diagnosis of cardiac malformation Pan et al.6). Results reported in presentation by Wellesley et al.6 years for males and 81. (70/422).5% QALY = quality-adjusted life year. Of these. giving the four chamber plus outflow tracts view a sensitivity for detecting TGA of only 20%.0 years for females.91 more live births per 1000 pregnancies having detected cardiac malformations antenatally (Table E.018 Wessex UK (1994–2005). 14 were only usually diagnosable with outflow-tract view. As shown in Table E.Diabetes in pregnancy Table E. the difference is almost entirely explained by the higher cost of the four chamber plus outflow tracts view ultrasound scan.86 neonatal deaths averted per 10 000 pregnancies (Table E.440 Pan et al.23 3.3 Results With baseline results. (1997) (see www.7).com/IDM/site/ idm4cr.92 Fetal echocardiography specificity 0.73 Source Smith et al.36 only detectable on four chamber plus outflow tracts view Fetal echocardiography sensitivity 0. (1997)275 Smith et al. EUROCAT.4 Outcomes and quality-adjusted life years Characteristic Life expectancy if TGA treated successfully (years) TGA mortality detected antenatally Value 76 Source ONS (2006) Notes UK life expectancy at birth (2003–05) is 76. Bonnet 1998–97418 1.418 Bonnet 1998–2002.440 Ritchie et al.00 Smith et al. (4/226).166 Wessex UK (1994–2005).a Four chamber plus outflow tracts 0. (1997)275 Ogge et al.3 Test characteristics Characteristic Four chamber view sensitivity Value 0. Following on from these cost and effects the estimated ICER for the four chamber plus outflow tracts view is £3. 194 . a Only one TGA was actually detected. (2004)416 TGA = transposition of the great arteries. E. EUROCAT. Assumes no long-term morbidity associated with successful TGA treatment.d4pro. TGA = transposition of the great arteries.95 Termination of pregnancy rate for 0. Discount rate stipulated by NICE guidelines manual 2007. the four chamber view is the cheapest strategy for screening for cardiac malformations.82 view sensitivity Four chamber plus outflow tracts 1.806 per QALY. A proportion of these (36% at baseline) would be TGA and given the baseline assumption about lower mortality for TGA with an antenatal diagnosis this leads to a concomitant 6. However.5.

985.000 £1.54 Live birth. but there must be ‘strong reasons’ for considering any intervention to be cost-effective if the cost per QALY is greater than £30.000 per QALY threshold used by NICE as a willingness to pay for cost-effectiveness.627 £2. The results of the sensitivity analyses are presented in Figures E.448 £2.344 £2. A £30.598 1. cardiac malformation detected Four chamber view Four chamber plus outflow tracts view 12.342 £2.806 is substantially below the £20.7. E. cardiac malformation not detected 10.250 £2.969 £2.784 £3.21 5.000.643 £2.5 Costs of four chamber and four chamber plus outflow tracts view strategies Costs Cardiac scan Fetal echocardiogram Termination of pregnancy Birth Total cost Cost per woman Four chamber view £34.Appendix E Table E.7 Incremental cost-effectiveness of four chamber plus outflow tracts view Incremental values Costs Antenatal diagnosis of cardiac malformations Antenatal diagnosis of TGA Neonatal deaths averted QALYs ICER Four chamber plus outflow tracts view £10.4 Sensitivity analysis A number of one-way sensitivity analyses were undertaken to assess to what extent uncertainty over certain parameter values was likely to be important in interpreting the baseline results.6 Outcomes of four chamber and four chamber plus outflow tracts view strategies Outcomes Pregnancy loss Termination of pregnancy Healthy live birth Live birth. Although the higher costs of the four chamber plus outflow tracts view make it more expensive than the four chamber view the ICER of £3.18 6.04 956.806 per QALY ICER = incremental cost-effectiveness ratio.91 0.995. NICE states that interventions with a cost per QALY of less than £20.000 cost per QALY threshold is indicated in each of the figures.940 £2. QALY = quality-adjusted life year. TGA = transposition of the great arteries.947.985 Four chamber plus outflow tracts view £46.37 956.000 should be considered costeffective.996 Table E. E.08 Table E.37 7.102 2.87 15.23 195 .5 Discussion With baseline values the model suggests that the four chamber plus outflow tracts view is costeffective for screening for cardiac malformations in pregnant women with diabetes compared with the four chamber view.87 17.686 0.2 to E.945.000 £1.47 12.

varying QALY weight of each baby treated for transposition of the great arteries (TGA) 196 .3 Incremental cost per QALY.2 Incremental cost per QALY. varying sensitivity of the four chamber view Figure E.4 Incremental cost per QALY. varying cost of the four chamber plus outflow tracts view Figure E.Diabetes in pregnancy Figure E.

2 suggests that the four chamber plus outflow tracts view would be cost-effective even if the test sensitivity for the four chamber view was within two percentage points of the sensitivity of the four chamber plus outflow tracts view. The model assumption is that these additional diagnoses are for malformations that would not normally be detectable with a four chamber view. as TGA is not the only malformation falling into this category. Figure E. the cost of the four chamber plus outflow tracts view would have to be £120 (for an incremental screening cost of £86) and well above the baseline estimate before the four chamber view would be preferred on cost-effectiveness grounds.13 incremental QALYs are needed to generate a cost per QALY of £30. The literature does not generally provide test sensitivity and specificity for individual cardiac malformations.Appendix E The model assumes that TGA is the only cardiac malformation where an antenatal diagnosis confers a benefit in terms of improved health outcomes for the woman and/or the baby.91 antenatal diagnoses of cardiac malformations using the four chamber plus outflow tracts view. Hence. instead it gives a value for detecting any cardiac malformation.806 in a population of 1000 pregnant women with diabetes. The one-way sensitivity analyses indicate thresholds for cost-effectiveness for other parameter values. the model does not assume that all additional antenatal diagnoses are TGA.000. Again it is the difference between screening costs using the four chamber view and the four chamber plus outflow tracts view that is important. The baseline parameter values give a TGA prevalence of approximately 4. The model follows the literature in using overall sensitivities and specificities and it is this which generates the additional 1. the improved sensitivity of the four chamber plus outflow tracts view compared with the four chamber view occurs because the four chamber plus outflow tracts view detects additional malformations that cannot usually be observed with the four chamber view (the sensitivity of detecting TGA with the four chamber view is 0%).686 TGA per 1000 pregnant women with diabetes. Baseline values suggest that the incremental costs of the four chamber plus outflow tracts view are £3. However. Therefore. only 0. which leads to the model result that a four chamber plus outflow tracts view would identify 0. 197 .5 show that the cost-effectiveness of the four chamber plus outflow tracts view is not sensitive to assumptions about QALYs or life expectancy within plausible ranges. With the baseline assumptions for perinatal mortality in relation to TGA detected antenatally or not.4 and E. it may be appropriate to assume a relatively low detection rate as published data reported that only one out of five TGA malformations was detected with a four chamber plus outflow tracts view. rather than the absolute costs of the screening tests. rather than the absolute values. It uses published data417 to estimate that 36% of the additional diagnoses would be TGA. The one-way sensitivity analysis of the sensitivity of the four chamber view is undertaken holding the sensitivity of the four chamber plus outflow tracts view constant at 82%.417 With the baseline detection rate used in the model it would be necessary to screen approximately 9800 pregnant women with diabetes using a four chamber plus outflow tracts view to avert one neonatal death. As the four chamber view sensitivity approaches that of the four chamber plus outflow tracts view there comes a point where there is only very limited added value in terms of detecting cardiac malformations using the four chamber plus outflow tracts view. although this is only 15% of the total TGA malformations present in the population. It should be noted that.6 per 1000 pregnancies in women with diabetes. The key issue here is the difference in test sensitivity between the four chamber view and the four chamber plus outflow tracts view.3 shows that the cost-effectiveness of the four chamber plus outflow tracts view compared with the four chamber view is quite sensitive to the costs of screening.3 QALYs per neonatal death averted. If a screening strategy involving a four chamber plus outflow tracts view detected all TGA malformations then the number of pregnant women with diabetes needed to screen with the four chamber plus outflow tracts view rather than the four chamber view to avert one neonatal death would be 1466. The baseline results suggest that the detection rate threshold for TGA for the four chamber plus outflow tracts view to achieve cost-effectiveness is quite low. Figure E. Figures E. However. but would be detectable with a view of the outflow tracts. one neonatal death would be averted for every seven TGA malformations detected. With baseline values this is approximately 1. the four chamber plus outflow tracts view still appears cost-effective with such a low detection rate. However. The sensitivity analysis suggests that the four chamber plus outflow tracts view requires a sensitivity that is at least four percentage points higher than the sensitivity for the four chamber view in order to achieve cost-effectiveness.

5 Incremental cost per QALY. varying mortality with diagnosis of transposition of the great arteries (TGA) Figure E.7 Incremental cost per QALY.Diabetes in pregnancy Figure E. varying the proportion of cardiac malformations only identifiable with four chamber plus outflow tracts view that are transposition of the great arteries (TGA) 198 . varying life expectancy Figure E6 Incremental cost per QALY.

8%) is comfortably below the threshold (14%) needed to yield a cost per QALY of £30.Appendix E Figure E. Given the way the model is constructed. TGA would have to account for less than 5% of the additional cardiac malformations detected for the ICER for four chamber plus outflow tracts view to exceed £30. 199 . Finally. this relates to the earlier discussion about the overall detection rate of TGA.000 per QALY. Clearly. Figure E. It does not address the ‘value’ or cost-effectiveness of such screening in providing information to inform a decision about termination of pregnancy. The results of the sensitivity analyses suggest that the cost-effectiveness of screening for cardiac malformations based on the four chamber plus outflow tracts view is robust and is unaffected by one-way variation of model parameters within plausible ranges. Here. the ‘best’ estimate used for baseline mortality for antenatally detected TGA mortality (1. a lower proportion of TGA malformations implies a lower detection rate. The model only addresses cost-effectiveness of screening for cardiac malformations in terms of the impact that antenatal diagnosis has on improved health outcomes. the cost-effectiveness of termination of pregnancy is problematic ethically and it does not readily fit into a QALY paradigm.000.7 shows that cost-effectiveness is sensitive to the proportion of additional cardiac malformations detected with the four chamber plus outflow tracts view that are assumed to be TGA. However. However.6 shows that the results of the model are sensitive to the assumptions made in relation to the positive impact that an antenatal diagnosis of TGA has on mortality.

Induction of Labour. London: NICE. How to use an article about therapy or prevention. 2003. 11. 14. et al. 17. JAMA: the journal of the American Medical Association 1994. London: NICE. London: CEMACH. Guidance on the Use of Patient-Education Models for Diabetes. London: RCOG Press. III. Are the results of the study valid? Evidence-Based Medicine Working Group.21(Suppl 2): B9–13. 24. 2nd ed. London: NICE. B. Diabetes in pregnancy: are we providing the best care? Findings of a national enquiry: England. British Medical Journal 1997. 2nd ed. 1994. 6. Series VS. III. National Institute for Clinical Excellence. NICE public health guidance 11. National Service Framework for Diabetes: Standards. Wales and Northern Ireland. Guyatt G and Sackett DL. How to use an article about a diagnostic test. Jaeschke R.315:275–8. No 32. National Institute for Health and Clinical Excellence. 3. Guyatt GH. 12.271(9):703–7. London: NICE. I. Wales and Northern Ireland. 18. 2006. London: RCOG Press on behalf of CEMACH. Torrance GW. 2004. How to get started. Drummond MF. National Institute for Health and Clinical Excellence. 2002. 1996. London: RCOG Press.270(17):2093–5.315:279–81. 3rd ed. 1999. Guidance on the Use of Continuous Subcutaneous Insulin Infusion for Diabetes. Sackett DL. II. London: NICE. Guideline Development Methods: Information for National Collaborating Centres and Guideline Developers. London: NICE. 13. 4. 2007. 23. Prospective population based survey of outcome of pregnancy in diabetic women: results of the Northern Diabetic Pregnancy Audit. 2004. Sackett DL and Guyatt GH. Type 1 Diabetes: Diagnosis and Management of Type 1 Diabetes in Children. Improving the Nutrition of Pregnant and Breastfeeding Mothers and Children in LowIncome Households. England. 31. London: NICE. 2007. Young People and Adults. Straus SE. How to practice and teach EBM. Methods for the economic evaluation of health care programmes. Scottish Intercollegiate Guidelines Network. London: Department of Health. 2005. 2002. Clinical Guidelines: Using Clinical Guidelines to Improve Patient Care Within the NHS. Geneva: World Health Organization. Evidence-based medicine. RCOG Clinical Effectiveness Support Unit. How to use an article about therapy or prevention. B. Edinburgh: Churchill Livingstone. London: NICE. National Insitute for Clinical Excellence. National Institute for Clinical Excellence. National Collaborating Centre for Primary Care. 20.271(5):389–91. Diabetes Care 1998. Evidence-based Clinical Guideline Number 9.271(1):59–63. How to use an article about a diagnostic test. Are the results of the study valid? Evidence-Based Medicine Working Group. National Collaborating Centre for Women’s and Children’s Health. British Medical Journal 1997. Richardson WS. Guidance on The Use of Glitazones for the Treatment of Type 2 Diabetes. 8. 200 . The Guidelines Manual 2007. 27. A. Users’ guides to the medical literature. 5. CEMACH. 22. Sackett DL and Cook DJ. Office for National Statistics. 32. Basingstoke: Palgrave Macmillan. 29. Oxman AD. Guyatt GH and Sackett DL. Intrapartum Care: Care of Healthy Women and Their Babies During Childbirth. diagnosis and classification of diabetes mellitus and its complications. National Institute for Clinical Excellence. Key Population and Vital Statistics 2005. Epidemiology of glucose intolerance and gestational diabetes in women of childbearing age. The Guidelines Manual 2006. CEMACH. Edinburgh: SIGN. Confidential Enquiry into Maternal and Child Health. JAMA: the journal of the American Medical Association 1994. 34. Sculpher M. 2006. 7. London: NICE [publication expected April 2008]. World Health Organization and Department of Noncommunicable Disease Surveillance. Part 1: diagnosis and classification of diabetes mellitus. 2004. JAMA: the journal of the American Medical Association 1994. Royal College of Obstetricians and Gynaecologists. 19. 2005. Jaeschke R. 10. London: NICE. 30. A guideline developers’ handbook. National Collaborating Centre for Women’s and Children’s Health. 2001. Oxford: Oxford University Press.270(21):2598–601. Casson IF. Antenatal Care: Routine Care for the Health Pregnant Woman (2008 Update). Users’ guides to the medical literature. Sackett DL and Cook DJ. Definition. Confidential enquiry into maternal and child health: Maternity services in 2002 for women with type 1 and type 2 diabetes. Postnatal Care: Routine Postnatal Care of Women and Their Babies. 2003. National Institute for Clinical Excellence. Report of a WHO consultation. et al. 26. 25. 2007. 2003. A. London: RCOG Press. No. 2. 9. Type 2 Diabetes: the Management of Type 2 Diabetes (Update). Users’ guides to the medical literature. 2001. 15. Guyatt GH. Users’ guides to the medical literature. Department of Health. 50. JAMA: the journal of the American Medical Association 1993. Guidance on the Use of Long-Acting Insulin Analogues for the Treatment of Diabetes – Insulin Glargine. Hawthorne G. National Institute for Health and Clinical Excellence. 28. No. NHS Executive. National Institute for Clinical Excellence. National Institute for Health and Clinical Excellence. What were the results and will they help me in caring for my patients? Evidence-Based Medicine Working Group. London: NICE. Confidential Enquiry into Maternal and Child Health: Pregnancy in Women with Type 1 and Type 2 Diabetes in 2002–03. The Evidence-Based Medicine Working Group. What are the results and will they help me in caring for my patients? The Evidence-Based Medicine Working Group. JAMA: the journal of the American Medical Association 1993. King H. London: HMSO. Local and Health Authority Areas. 2007. 21. 33. London: RCOG Press. Caesarean Section. 2008.References 1. Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study. Users’ guides to the medical literature. National Collaborating Centre for Women’s and Children’s Health. 2005. London: CEMACH. 2008. II. 2000.

Spontaneous abortion in patients with insulin-dependent diabetes mellitus: the effect of preconceptional diabetic control. DAFNE Study Group.174(4):1343–53. The influence of maternal weight and glucose tolerance on infant birthweight in Latino mother-infant pairs. Gavin LA.277:74–6. Low energy diets in the treatment of gestational diabetes. Driscoll SG. Northern Ireland. Home and Health Department. et al. 62. Iqbal S. Cloherty JP.301(6760):1070–4. Singer MR.84(4):515–20. Miodovnik M. In: Cochrane Database of Systematic Reviews. Summerbell C. Hare JW. Epidemiology 2000. Cloherty JP. 40. Diabetes Care 2004. Department of Health and Social Services. Exercise for diabetic pregnant women. Moelsted-Pedersen L. 68. 53. Holroyde JC. Carman WJ. Di Renzo GC. Diabetes Care 2003. Miodovnik M. London: Taylor & Francis Group. In: Hod M. Obstetrics and Gynecology 1986. 47.27(12):2819–23. Obstetrics and Gynecology 1991.157(6):1460]. Petocz P. 41. Outcomes in type 1 diabetic pregnancies: a nationwide. Experimental and Clinical Endocrinology 1984. et al. American Journal of Obstetrics and Gynecology 1987. eds. Miodovnik M. 65. Supplementum 1986. et al. 60. Aula P.39(3):225–31.158(5):1161–4.26(8):2261–7. Gillmer MD.319(25):1617–23. Glycemic control prevents congenital anomalies. Jovanovic L. population-based study. Leveno KJ. Steel JM. Diabetologia 2000. (Cochrane Review). 37.94(7):347–56. Reiher H. Chichester: Wiley Interscience. Little J. Skillman C. Diabetes Care 1998. et al. Brand-Miller J. Bradlee ML. 45. First-trimester hemoglobin A1 and risk for major malformation and spontaneous abortion in diabetic pregnancy. et al. Beard RW. Can prepregnancy care of diabetic women reduce the risk of abnormal babies? British Medical Journal 1990. Moore LL. Treatment of pregnant insulin-dependent diabetic women. 59. QJM: monthly journal of the Association of Physicians 2001. Goldman JA. Training in flexible. Dietary advice for the prevention of type 2 diabetes mellitus in adults. Feldberg D. The Diabetes Control and Complications Trial Research Group. The effect of intensified conventional insulin therapy before and during pregnancy on the malformation rate in offspring of diabetic mothers. Vermeulen MJ. Suhonen L. Hooper L. Ylinen K. Lucas MJ. Johnstone FD. Rosenn B. Lopez J. 2003.References 35. Dicker D. Pre-conception management of insulin-dependent diabetes: improvement of pregnancy outcome. Greene MF. 56. 57. 54. Diabetes and Pregnancy Group F. Expert Advisory Group. Mozas J. Ceysens G. Samuel N. Predictive value of early pregnancy glycohemoglobin in the insulin-treated diabetic patient. 43. 64. (Cochrane Review). Glycaemic control during early pregnancy and fetal malformations in women with type I diabetes mellitus. et al. Early pregnancy glycosylated hemoglobin.155(2):293–7. Reilly R. 67. Miodovnik M. Rosenn B. et al. 340–58. 1992. Mills JL. 63. Hare JW.153(4):439–42. Gestational weight gain and adverse neonatal outcome among term infants. Damm P. Semmler K. American Journal of Obstetrics and Gynecology 1996. Gold AE.265(6):731–6.277:44–9. Diabetologia 2005. 55. 61. 46. Rouiller D and Boulvain M.21(4):535–8.304(22):1331–4.68(3):366–9. Incidence of spontaneous abortion among normal women and insulin-dependent diabetic women whose pregnancies were identified within 21 days of conception.11(6):689–94. et al. Dornhorst A and Frost G. Obstetrics and Gynecology 1994. Diabetes and Pregnancy. 66.83(2):173–7. Welsh Office. Nutritional management in diabetic pregnancy: a time for reason not dogma. Stenman UH. et al. Maternal and neonatal outcomes in pregestational and gestational diabetes mellitus. 58. England BG. et al. severity of diabetes.108(3 Pt 1):635–43. et al.26(11):2990–3. Glycemic control and spontaneous abortion in insulin-dependent diabetic women. 201 . Hopkins LM. et al. JAMA: the journal of the American Medical Association 1991. et al. Holcberg G. Glycemic thresholds for spontaneous abortion and congenital malformations in insulin-dependent diabetes mellitus. Jensen DM. Low-glycemic index diets in the management of diabetes: a meta-analysis of randomized controlled trials. Kieffer EC. Maresh M. 39. British Medical Journal 1984. Cheng YW. Rosenn BM. Hayne S. et al. et al. Scottish Office. Is pregnancy in diabetic women associated with folate deficiency? Diabetes Care 1999. Preconception care of diabetes. British Medical Journal 2002. et al. 52. et al. Key TC. Kaplan JS. Giuffrida R and Moore TR. Acta Endocrinologica Supplementum 1986. Diabetes Care 2003.325:746–8.96(12):2201–8.85(3):417–22. et al.22(7):1017–21. et al. Acta Endocrinologica. Department of Health. Obstetrics and Gynecology 1995. American Journal of Obstetrics and Gynecology 1985. intensive insulin management to enable dietary freedom in people with type 1 diabetes: dose adjustment for normal eating (DAFNE) randomised controlled trial. Mimouni F. Fuhrmann K. Langer O. and the influence of maternal obesity and weight gain: The DEPOSIT study. 50. 42. et al. Miodovnik M. Miller E. Obstetrics and Gynecology 2006. Hiilesmaa V and Teramo K.43(1):79–82. Stotland NE. Folic acid and the prevention of neural tube defects. [Erratum appears in Am J Obstet Gynecol 1987. 2005. Hepburn DA. Risk of minor and major fetal malformations in diabetics with high haemoglobin A1c values in early pregnancy. p. Elevated maternal hemoglobin A1c in early pregnancy and major congenital anomalies in infants of diabetic mothers. A prospective study of the risk of congenital defects associated with maternal obesity and diabetes mellitus. Elevated maternal glycohemoglobin in early pregnancy and spontaneous abortion among insulin-dependent diabetic women. Hypoglycemia: the price of intensive insulin therapy for pregnant women with insulindependent diabetes mellitus. French multicentric survey of outcome of pregnancy in women with pregestational diabetes. Tabaei BP. et al. American Journal of Obstetrics and Gynecology 1989. et al. de Leiva A. New England Journal of Medicine 1981. and fetal malformations. 49.77(6):846–9. Tsang RC. 44. 48. American Journal of Obstetrics and Gynecology 1988. et al. Teratology 1989. Williams ML. Ricart W. 36.161(2):426–31. Combs CA. American Journal of Public Health 2006. Pregnancy outcomes in the Diabetes Control and Complications Trial. 2006. et al. Chichester: Wiley Interscience. Simpson JL. American Journal of Obstetrics and Gynecology 1986. Pregnancy outcome in patients with insulin-dependent diabetes mellitus with preconceptional diabetic control: a comparative study. In: Cochrane Database of Systematic Reviews. Ray JG.156(5):1096–100.48(9):1736–42. 69. 38. Fuhrmann K. Combs CA. The effect of glycemic control in the pre-conception period and early pregnancy on birth weight in women with IDDM. New England Journal of Medicine 1988. Kitzmiller JL. et al. Body mass index has a greater impact on pregnancy outcomes than gestational hyperglycaemia. Feldberg D. 51. et al. Shapiro JL. Gin GD.289(6441):345–6. Dicker D. Moore H.

9(1):66–9. 2005.20(9):779–80. et al. Berkus MD. Kozer E.22(6):803–7. New England Journal of Medicine 2000. 105. 92.350(9089):1446–7. McCance D. A comparison of glyburide and insulin in women with gestational diabetes mellitus. Diabetes Care 2005.55(Suppl 1):A417–1804-P (Abstract). 91. Diabetes Care 2003. 359–78. 2007. Diabetic nephropathy and pregnancy: the effect of ACE inhibitors prior to pregnancy on fetomaternal outcome. Insulin therapy in pregnancy. et al. Transfer of insulin lispro across the human placenta: In vitro perfusion studies. van Loon AJ. 82. Insulin glargine use in pregnancy is not associated with adverse maternal or fetal outcomes.329(14):977–86. Diabetes 2006. et al. Insulin aspart vs. 84. Volpe L. Diabetic Medicine 2005. Swahn ML. 98. p. Pregnancy and the long-acting insulin analogue: A case study. Howard C. Philadelphia: Lippincott.23(3):223–7. Poyhonen-Alho M. Valois M and Koren G. The utility and efficacy of the new insulins in the management of diabetes and pregnancy. 93. et al. 202 . Textbook of Diabetes and Pregnancy. 73. Anil S. Mecacci F. Bartlett C and Gillmer MD. Diabetes Research and Clinical Practice 2002. Karp M. Contraindications to use of metformin. Bar J.48(Suppl 1):A317. et al. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. Insulin glargine during pregnancy. eds. Gynecology and Reproductive Biology 2003. 85. Diabetic Medicine 2003. London: British Medical Association and Royal Pharmaceutical Society of Great Britain. Freeman RK and Yaffe SJ. Hernandez-Diaz S. Cypryk K. Diabetes 2005. Diabetic Medicine 2007. Frias JL.354(23):2443–51.343(16):1134–8. 90. 78. Pettitt D. Diabetes 2005.114(4):453–7. Current Diabetes Reports 2002. Balaji V and Seshiah V. 104. Jovanovic L.24(3):253–8. 79. 89. 102. Use of insulin glargine during pregnancy: A case–control pilot study. European Journal of Obstetrics. 87. 76. et al. Use of insulin glargine during the first weeks of pregnancy in five type 1 diabetic women. 72.77(1):24–6. Pregnancy outcome in patients with insulin dependent diabetes mellitus and diabetic nephropathy treated with ACE inhibitors before pregnancy.111(1):19–24.20(1):46–50.58(2):115–21. Acta Bio-Medica de l Ateneo Parmense 2006. Hoyme HE. Ekpebegh CO.55(Suppl 1):A40. Simmons D. Hjertberg R. Diabetes Care 2006.28(4):982–3. Hawthorne G. Mathiesen E. Di Renzo GC. Langer O. Brash PD. London: Martin Dunitz. Congenital anomaly rate in offspring of mothers with diabetes treated with insulin lispro during pregnancy. 77. A 10-year retrospective analysis of pregnancy outcome in pregestational Type 2 diabetes: comparison of insulin and oral glucose-lowering agents. Magee LA.54(Suppl 1):A787 (Abstract). Diabetes 2006. 88.55(Suppl 1):A417 (Abstract). Berghold A.10(2):PI29–32. Use of insulin glargine during pregnancy in seven type 1 diabetic women. Di CG. Jovanovic L. 97.Diabetes in pregnancy 70. Williams and Wilkins.12(5):659–65. et al. 95. Gallen IW and Jaap AJ. 80. Masson EA. Journal of Pediatric Endocrinology 1999. Lip GY. In: Hod M. Briggs GG. Archives of Internal Medicine 2005. van der ML. 71. et al. Al-Agha R. Feig DS. Holstein A. et al. New England Journal of Medicine 1993. Elder AT. Ronnemaa T. 75. et al.2(4):331–6. Storms F. Graves DE. BJOG: an international journal of obstetrics and gynaecology 2007. A Reference Guide to Fetal and Neonatal Risk. Al-Shaikh AA.10(12):2328–33. Carignani L. 96. Pregnant women with type 1 diabetes mellitus treated by glargine insulin. Nephrology Dialysis Transplantation 1995. Frias JP. Churchill D. Insulin aspart – safe during pregnancy.28(10):2594–5. Arbogast PG. Endocrine Practice 2003. 53rd ed. Maternal metabolic control and perinatal outcome in women with gestational diabetes treated with regular or lispro insulin: comparison with non-diabetic pregnant women. et al. et al.326(7392):762. Persson B. Pregnancy outcome in Type 1 diabetes mellitus treated with insulin lispro (Humalog). Use of insulin glargine during embryogenesis in a pregnant woman with Type 1 diabetes. Derewlany L. Angiotensin-converting-enzyme inhibitors in early pregnancy. 106.86(3):658–63. Lancet 1997. Insulin lispro therapy in pregnancies complicated by type 1 diabetes: glycemic control and maternal and fetal outcomes.54(Suppl 1):A461 (Abstract). Kitzmiller JL and Jovanovic L. et al R. et al. Schoenfeld A. et al. Hod M. 100. 74.165(12):1337–44. New England Journal of Medicine 2006. Kinsley B. Boskovic R. Cioni F. British National Formulary. 7th ed. Diabetes 2006. Fertility and Sterility 2006. et al. Diabetes Control and Complications Trial (DCCT) Research Group. Systematic review and meta-analysis of short-acting insulin analogues in patients with diabetes mellitus. Pregnancy complications and perinatal outcome in diabetic women treated with Humalog (insulin lispro) or regular human insulin during pregnancy. 62. 81. 101. Sobczak M. et al. Metformin use and diabetic pregnancy – has its time come? Diabetic Medicine 2006. Saudi Medical Journal 2006. Murray S. de Leiva A. Diabetes Care 2005. Pregnancy outcome after first-trimester exposure to metformin: a meta-analysis. 86. White JC and Kirk JK. Plank J. Age and creatinine clearance need to be taken into consideration. et al. Coetzee EJ. et al. Langer O. Patmore JE. Wyatt JW. 103. Woolderink JM. Journal of Maternal-Fetal Medicine 2000.27(4):563–5. Diabetic Medicine 2003. 83. Chen R.10(4):179–83. Drugs in Pregnancy and Lactation. The use of insulin glargine with gestational diabetes mellitus.29(2):471–2. Joint Formulary Committee. Caronna S. A comparison of soluble human insulin vs rapid acting insulin analogue in type 1 diabetes mellitus in pregnancy. Saltevo J.9(3):187–93. et al. Insulin lispro therapy in pregnancies complicated by type 1 diabetes mellitus. Cioni R. Conway DL. Garg SK. Canadian Journal of Clinical Pharmacology 2003. Major congenital malformations after first-trimester exposure to ACE inhibitors. regular human insulin in basal/bolus therapy for patients with gestational diabetes mellitus: safety and efficacy.26(5):1390–4. 99. British Medical Journal 2003. Medical Science Monitor 2004. Pertynska-Marczewska M. Beevers M. 2003. Siebenhofer A. van Dijk DJ. Plaschke A and Egberts EH. Gilbert C. Kinsley BT. Cooper WO. Diabetologia 2005. Price N. Selecting antihypertensive therapy in the pregnant woman with diabetes mellitus. et al. et al. Conway DL and Longer O. 94. Lencioni C. Gutzin SJ. Maternal hyperglycemia and glycemic control in pregnancy: a randomized trial comparing insulin aspart with human insulin in 322 subjects with type 1 diabetes. The safety of oral hypoglycemic agents in the first trimester of pregnancy: A meta-analysis. Dall’Aglio E.

115. The safety of calcium channel blockers in human pregnancy: a prospective. et al.42(9):1036–42. American Diabetes Association. et al. Charron-Prochownik D. Petersen S. New England Journal of Medicine 1986. Becker MP. Griffiths R and Davis W. Punnose J. Knowledge. Spong CY. Unplanned pregnancies in young women with diabetes. Johnson H. and Israel. et al. Factors associated with seeking pre-conception care. Cheung NW. attitudes and behaviors related to sexuality and family planning in adolescent women with and without diabetes. Diabetes and pregnancy.3(4):602–13. 113. 143.278(13):1078–83. Carey VJ. Buckley C. Dornhorst A. Pregnancy outcome with ACE-inhibitor use in early pregnancy. Magann EF. et al.9:820–5.73(4):557–61. Scott DA. Rabinerson D and Peled Y. Gestational diabetes: dilemma caused by multiple international diagnostic criteria. 123.17(1):26–32.55(4):545–57. London: British Heart Foundation. Obstetrics and Gynecology 1990. Berger DK and Chamany S. Giampietro O. Agarwal MM. MMWR . Francis J. Diabetic Medicine 2005. Moses R. Weeks J. Younger MD. et al. Bekedam DJ.69(3):279–86. 109. 137. 134. et al. Angiotensin converting enzyme inhibitors use in the first trimester of pregnancy. Sorensen HT. 119. 1987–1995. 203 . et al. 136. Recurrence of gestational diabetes mellitus: identification of risk factors.Morbidity and Mortality Weekly Report 1997. Preconception care and the risk of congenital anomalies in the offspring of women with diabetes mellitus: A meta-analysis. et al. Holing EV. American Journal of Perinatology 1998.174(3):823–8. 120. Diabetes Care 2004. Davey RX and Hamblin PS. et al. Jorgensen LG. Clarke P. et al. Loveman E. Schytte T. Barrett G and Wellings K. The clinical usefulness of glucose tolerance testing in gestational diabetes to predict early postpartum diabetes mellitus. St James PJ. et al. McIntyre L.22(12):1731–6. Coronary Heart Disease Statistics: Diabetes Supplement. Diabetic Medicine 2005. Casele HL and Laifer SA. 128. Weijers RN. Cleave B and Tomlinson G. Relation of glucose tolerance to complications of pregnancy in nondiabetic women.72(1):105–20. 125.16(12):1572–8. 116. Ostlund I and Hanson U. JAMA: the journal of the American Medical Association 1997. Acta Obstetricia et Gynecologica Scandinavica 2003. Doherty DA. 122. 132. Acta Endocrinologica 1954.27(Suppl 1):S5–10. Postmarketing surveillance for angiotensin-converting enzyme inhibitor use during the first trimester of pregnancy--United States. Diagnosis and classification of diabetes mellitus. Health Technology Assessment 2002.30(5):1314–19. 121. 112. Dhatt GS. Sereika SM. 141. Quarterly Journal of Medicine 2001. O’Brien TE and Chan WS.16(4):330–42. Rayner M. Pediatric Diabetes 2006. Diabetic Medicine 1992. Hod M and Merlob P. The clinical impact of screening for gestational diabetes. Griffin ME. 114.315(16):989–92. Archives of Internal Medicine 1998. Bar J. Medical Journal of Australia 2001. Diabetes Educator 2006.44(1):99–104. Hamilton BD. Keshavarz M. Pre-pregnancy body mass index and pregnancy outcomes. Recurrence of gestational diabetes: who is at risk? American Journal of Obstetrics and Gynecology 1998. risk factor-based screening for gestational diabetes mellitus: detection rates. An analysis of psychosocial factors. et al. Babaee GR. Carpenter MW. 127. Paterson CM. The plans in unplanned pregnancy: Secondary gain and the partnership. Harris K and Campbell E. Nicholls JSD. Lancet 1998. et al. Obstetrics and Gynecology 1989.22(4):507–8. Diabetes Care 2007. 2001. 124. et al. Charron-Prochownik D. 111. Clinical Chemistry and Laboratory Medicine 2006. et al.179(4):1038–42. Social Science and Medicine 2002.18(2):157–65. A prospective study of pregravid determinants of gestational diabetes mellitus. Canada.46(11):240–2. Goldschmidt HM. Coleman MA. Buccimazza A. International Journal of Risk and Safety in Medicine 1997. 135. Nielsen GL. 133. et al. Willett WC.References 107. 139. DeVeciana M.6(11):1–172. Recurrence of gestational diabetes mellitus: a systematic review. Tallarigo L. Gestational diabetes: do all women need to be tested? Australian and New Zealand Journal of Obstetrics and Gynaecology 1995. 117. Norman P. Coustan DR. Diabetes Research and Clinical Practice 2005. 130. et al. Penno G. Beyer CS. International Journal of Gynecology and Obstetrics 2006. Brown ZA. et al. Weight and length at birth of infants of diabetic mothers.32(2):235–42. Pedersen J. Diabetes Care 1993.29(3):526–30. Long-term effects of a diabetes and pregnancy program: does the education last? Diabetes Care 2006.35(4):387–9. Occurrence of gestational diabetes mellitus and the value of different screening indicators for the oral glucose tolerance test. 131. Diabetes Care 1995. Selective versus universal screening for gestational diabetes mellitus: an evaluation of predictive risk factors. Donnenfeld AE.10(1):23–6. Reproductive health and preconception counseling awareness in adolescents with diabetes: what they don’t know can hurt them. The introduction of a specific request form for the diagnosis of gestational diabetes (GDM) improves understanding of GDM amongst clinicians but does not increase its detection.75(6):970–4. Factors influencing preconception control of glycemia in diabetic women. Steffensen FH.94(8):435–44. multicenter cohort study. 118. Gestational diabetes mellitus: Is it a clinical entity? Diabetes Reviews 1995. High prevalence of gestational diabetes in women from ethnic minority groups. Kuboshige J. et al. Maternal age and screening for gestational diabetes: A population-based study. Diabetic Medicine 2000.15(1):29–33. Janz NK. Hod M. et al. Coffey M. Ray JG. Screening for gestational diabetes: a systematic review and economic evaluation. Magee LA. Major CA. Solomon CG. 110.174(3):118–21. Centers for Disease Control and Prevention (CDC). 138. Anthony J. 140.158(12):1321–4. Brandslund I. 129. Nelson C.82(2):103–8. Wang SL. Universal vs. Sereika SM. 126. Why don’t women with diabetes plan their pregnancies? Diabetes Care 1998. risk factors and pregnancy outcomes. Falsetti D.351(9102):596. British Journal of Medical Psychology 1999. Kim C. Schick B. Belfort MA.95(3):242–7. What is a ‘planned’ pregnancy? Empirical data from a British study.21(6):889–95. et al. 108. et al. Herman WH.7(5):267–73. Hemodynamic changes associated with intravenous infusion of the calcium antagonist verapamil in the treatment of severe gestational proteinuric hypertension. 142. Clinical Chemistry and Laboratory Medicine 2004. gestation at diagnosis and outcome. Gestational diabetes in Iran: incidence. American Journal of Obstetrics and Gynecology 1996. Feig DS. Guillermo L. et al.

Naylor CD. European Journal of Clinical Nutrition 2006. American Journal of Obstetrics and Gynecology 1989. Terry RB.21(Suppl 2):B33–42.95(3):223–9. Durak EP and Peterson CM. Journal of the American College of Nutrition 1995.65(4):487–91.17(4):275–83.15(2):145–56. de Veciana M. Improved glucose tolerance in gestational diabetic women on a low fat. 170. Gunnells DJ. Sacks DA. A randomized controlled trial using glycemic plus fetal ultrasound parameters versus glycemic parameters to determine insulin therapy in gestational diabetes with fasting hyperglycemia. Maternal carbohydrate intake and pregnancy outcome. Di Giovanni BM. 175. Calorie restriction for treatment of gestational diabetes. Proceedings of the Nutrition Society 2002. Peterson CM and Jovanovic-Peterson L. Ryan EA. Diabetes 1991.61(1):45–50. 146.26(4):1206–10. Brankston GN. Botta RM. Diabetes Care 2006. high unrefined carbohydrate diet. Sorensen B. Kaufmann RC.40(Suppl 2):161–4. 155. American Journal of Obstetrics and Gynecology 1987. Early third-trimester ultrasound screening in gestational diabetes to determine the risk of macrosomia and labor dystocia at term. Chen W. Obstetrics and Gynecology 1985. Shragg P and Hollingsworth DR. Morgan MA. 156. Anyaegbunam A. A preliminary review. Diabetes Care 2003. Hiller JE. et al. Prevention of neonatal macrosomia in gestational diabetes by the use of intensive dietary therapy and home glucose monitoring. et al. 159. Insulin sensitivity in third trimester pregnancy. Tapsell LC. Hansson U. et al. Thompson DJ. Screening or diagnostic: markedly elevated glucose loading test and perinatal outcomes. Diabetes Care 1998. Journal of Maternal-Fetal Medicine 1996. Moss JR. Can glucose tolerance test predict fetal hyperinsulinism? BJOG: an international journal of obstetrics and gynaecology 2000. 166. 176.190(1):188–93. 172. 152. Evans S. et al. et al. Jensen DM.107(12):1480–5. Proposed diagnostic thresholds for gestational diabetes mellitus according to a 75-g oral glucose tolerance test. Tamussino K. Australian and New Zealand Journal of Obstetrics and Gynaecology 1984. Sacks DA. The association between impaired glucose tolerance and birth weight among black and white women in central North Carolina. American Journal of Obstetrics and Gynecology 2004. Rae A. 163. Porter KB. Frid AH. Postprandial versus preprandial blood glucose monitoring in women with gestational diabetes mellitus requiring insulin therapy. 164. British Journal of Obstetrics and Gynaecology 1988. Cammilleri F. 171. 154. 145.157(3):703–8. 178.181(4):904–11. Groop LC. 148.14(4):369–75. Hanson U. Farine D. et al.24(11):1904–10. 177. 204 . The Toronto Tri-Hospital Gestational Diabetes Project. Dornhorst A. et al. Persson B. 174. Gestational diabetes mellitus (GDM). Dornhorst A and Frost G. Diabetes Care 1998. Gillen L. 150. et al. Probst F. A dietary exchange of common bread for tailored bread of low glycaemic index and rich in dietary fibre improved insulin economy in young women with impaired glucose tolerance. et al.5(4):211–17. Amankwah KS.59(2):135–43. et al. Ford FA and Lawrence GF. 153.34(Suppl 2):101–4. 169. 173. Diabetes Care 2003. Effect of dietary carbohydrate on the glucose and insulin response to mixed caloric intake and exercise in both nonpregnant and pregnant women. Lesser KB. Diabetes Care 2003. Journal of Human Nutrition and Dietetics 2002. Haeusler M. 165. Schaefer-Graf U. et al. Sermer M. The 75-gram glucose load in pregnancy: relation between glucose levels and anthropometric characteristics of infants born to women with normal glucose metabolism. Sardesi S. Cioni R. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. 149. American Journal of Obstetrics and Gynecology 1995. Block-Kurbisch I. et al. The significance of one abnormal glucose tolerance test value on adverse outcome in pregnancy. diet versus insulin and diet. Esakoff TF. Damm P. 157. et al. When is fasting really fasting? The influence of time of day. et al. et al. Bjorklund A. East African Medical Journal 1981. et al. Buchanan TA. 151. Understanding exercise beliefs and behaviors in women with gestational diabetes mellitus. Diabetes Care 1994. Parretti E. et al. Medearis AL. Diabetes Care 2001. Annali Dell’Istituto Superiore di Sanita 1997. A randomized study of dietary effects. Algert S. Toward universal criteria for gestational diabetes: the 75-gram glucose tolerance test in pregnancy.161(2):415–19. et al.21(Suppl 2):B107–12.172(2 Pt 1):607–14. 158. The Australian and New Zealand Journal of Obstetrics and Gynaecology 2000.26(7):2107–11. Nicholls JS. Randomized trial of diet versus diet plus cardiovascular conditioning on glucose levels in gestational diabetes. Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS) Trial Group. Brustman L. et al. Use of fetal ultrasound to select metabolic therapy for pregnancies complicated by mild gestational diabetes. Maternal and fetal outcomes if gestational impaired glucose tolerance is not treated. 162. 160. American Journal of Obstetrics and Gynecology 1987. New England Journal of Medicine 2005. Wechter DJ. Saldana TM.157(3):758–63. Clapp JF 3rd. Mitchell BF. et al. et al. Obstetrics and Gynecology 1990. American Journal of Perinatology 1991. Journal of Maternal-Fetal and Neonatal Medicine 2006.58(2):90–4. bu-Fadil S.33(3):403–6. Wolde-Tsadik G.60(3):334–41. Mello G. et al. Bond D. et al. Adair LS.20(1):51–7. The type and frequency of consumption of carbohydrate-rich foods may play a role in the clinical expression of insulin resistance during pregnancy.19(11):729–34. Nutrition and Dietetics: Journal of the Dietitians Association of Australia 2002. Bochner CJ.8(2):131–4. A randomised controlled trial of dietary energy restriction in the management of obese women with gestational diabetes. Major CA. Maternal and perinatal outcomes in 3260 Danish women. Diabetes 1985.Diabetes in pregnancy 144. Ostman EM.26(3):656–61. Montoro MN. 147. 55% carbohydrate weight loss strategies in women with previous gestational diabetes mellitus and non-diabetic women of 130–200% ideal body weight. Martin GS.29(2):236–40. Nolan CJ. Symons DD and Ulbrecht JS.352(24):2477–86. Prophylactic insulin in the management of gestational diabetes. New England Journal of Medicine 1995. The effect of pregnancy on the normal range of the oral glucose tolerance in Africans. The principles of dietary management of gestational diabetes: reflection on current evidence. Moderate caloric restriction in obese women with gestational diabetes. Crowther CA. Gruppuso PA. Kjos SL. Fraser RB. Clapp JF 3rd. Predictive factors for insulin treatment in women with diagnosis of gestational diabetes. Diabetic Medicine 2003. Williams J 3rd. Greenspoon JS. Ostlund I. 161.40(4):416–22. Kjos SL. Jovanovic-Peterson L. Randomized crossover study of 40% vs. Stangenberg M. et al. Fraser RB. Comparative evaluation of two treatment regimens. Langer O.333(19):1237–41. American Journal of Obstetrics and Gynecology 1999. 167.24(3):174–7. Cheng YW. Resistance exercise decreases the need for insulin in overweight women with gestational diabetes mellitus. interval after a meal. and maternal body mass on maternal glycemia in gestational diabetes. et al..75(6):960–4. Weiss PAM. Siega-Riz AM. 168. Exercise fails to improve postprandial glycemic excursion in women with gestational diabetes.

Kremer CJ and Duff P. Minor congenital malformations in infants of insulin-diabetic women: association with poor glycaemic control. di Biase N. Di Renzo GC. Acta Obstetricia et Gynecologica Scandinavica 2002. 209. et al. Florida. Continuous glucose monitoring for the evaluation of gravid women with type 1 diabetes mellitus. Mol BWJ. Napoli A. American Journal of Obstetrics and Gynecology 1972.or long-acting insulin and neonatal outcome: a pilot study. Bonomo M. Perinatal outcomes and the use of oral hypoglycemic agents. 207. The National Institute of Child Health and Human Developmen – Diabetes in Early Pregnancy Study.164(1 Pt 1):103–11.33(3):347–51. Karlsson K and Kjellmer I. and regular human insulin with no insulin in gestational diabetes mellitus. Effect of euglycemia on the outcome of pregnancy in insulin-dependent diabetic women as compared with normal control subjects. 199. The outcome of diabetic pregnancies in relation to the mother’s blood sugar level. Papini M. et al. American Journal of Obstetrics and Gynecology 1994. Kitzmiller JL. Paulk TH and Kee CC. 183. Conway DL. Metabolic and immunologic effects of insulin lispro in gestational diabetes. 210.30(3):237–44. et al. 180. Preprandial versus postprandial blood glucose monitoring in type 1 diabetic pregnancy: a randomized controlled clinical trial. insulin aspart. Gonzales O and Skiver D.10(4):241–5. Xenakis EM. 187. Somigliana E. Kamitomo M. Combs CA. Diabetic Medicine 1999. Scientific abstracts of the seventh Annual Meeting of the Society for Perinatal Obstretrics Lake Buena Vista. Factors influencing birth weight in infants of diabetic mothers. Diabetes Care 2004. et al. Acta Obstetricia et Gynecologica Scandinavica 2000. Donnan PT. Teramo K and Kaaja R. Obstetrics and Gynecology 2003. Evers IM. Bertini AM. Intensified versus conventional management of gestational diabetes.101(4):633–8. Chen R. Kajiya S. 195. Druzin M and Peterson CM. et al.190(5):1438–9. 188. Ilic S.71(6):921–7.189(2):507–12. Gunderson E.21(2):129–34. et al. et al. Hod M.170(4):1036–46. et al. Moller M and Sorensen HT. Pisoni MP. Jovanovic L. American Journal of Obstetrics and Gynecology 1991. Ospina P.15(10):1251–7. Valuk J. Diabetes Care 2003.45(11):1484–9. Diabetes 1986. Peterson CM. Diabetes Care 2001.24(8):1319–23. 5–7 February 1987 (Abstract). Flexible treatment of gestational diabetes modulated on ultrasound evaluation of intrauterine growth: a controlled randomized clinical trial. Ching JY. Poor glycated haemoglobin control and adverse pregnancy outcomes in type 1 and type 2 diabetes mellitus: Systematic review of observational studies. et al. Miodovnik M. Hadden DR. Schaefer-Graf UM.104(1):88–93. Pettitt DJ. 197. et al. Maternal postprandial glucose levels and infant birth weight: the Diabetes in Early Pregnancy Study. Third-trimester maternal glucose levels from diurnal profiles in nondiabetic pregnancies: correlation with sonographic parameters of fetal growth. 212. Rossi G. Smits MW. Nielsen GL. American Journal of Obstetrics and Gynecology 2005. A randomized trial evaluating a predominantly fetal growth-based strategy to guide management of gestational diabetes in Caucasian women. Assessing the impact of an outpatient education program for patients with gestational diabetes. Inkster ME. Fauzan OH. Jacobson GF. De Valk HW and Visser GH. Cetin I. HbA1c in early diabetic pregnancy and pregnancy outcomes: A Danish population-based cohort study of 573 pregnancies in women with type 1 diabetes. Diabetes Care 1999. Ramos GA. The role of continuous glucose monitoring in gestational diabetes mellitus. Annali Dell’Istituto Superiore di Sanita 1997. 206. 191. Insulin-meal interval and short-term glucose fluctuation in tightly controlled gestational diabetes mellitus.11(4):260–2. London: Martin Dunitz.81(3):258–9. 198. Results from a randomized study. 192. Jovanovic L. Does tight control of blood glucose in pregnant women with diabetes improve neonatal outcomes?.49(1):25–8. 211. 193. Comparison of glyburide and insulin for the management of gestational diabetes in a large managed care organization. Taborda W. 205. fetal macrosomia. Relationship of fetal macrosomia to maternal postprandial glucose control during pregnancy. Jovanovic L. et al. Obstetrics and Gynecology 1990. Kjos SL.112(2):213–20. American Journal of Medicine 1981. Williams FL and Harper V. Wyse LJ. Diabetes Educator 1995. Rosenn B. Textbook of Diabetes and Pregnancy. Diabetes and Metabolism 2004. Neonatal morbidity in pregnancy complicated by diabetes mellitus: predictive value of maternal glycemic profiles. Patterson CC.156(5):1089–95. Glyburide for the treatment of gestational diabetes. American Journal of Obstetrics and Gynecology 1987. Use of glyburide for the treatment of gestational diabetes: the San Antonio experience. Journal of Family Practice 2004.26(1):183–6. 196. et al. Silva JC. 205 . Kolaczynski JW. 2003.References 179.27(2):297–302. et al. Screening practices for gestational diabetes mellitus in UK obstetric units. 194. Piana R.76:745–9. Ben-Haroush A. 201.22(9):1422–7. Landon MB. Manderson JG. Gabbe SG. Ben-Haroush A. et al.29(12):2612–16. 186. Diabetes Care 2006. results of a nationwide study in The Netherlands. et al. 213. De Valk HW.193(1):118–24. et al. insulin. Comparison of an insulin analog. Sabbatini A.79(8):649–54. Adequate timing of fetal ultrasound to guide metabolic therapy in mild gestational diabetes mellitus. The Journal of Maternal-Fetal Medicine 2001. Fahey TP. Jovanovic-Peterson L. Jovanovic L. Treatment of gestational diabetes with short. Diabetes Care 1992. 184. 204. Pettitt DJ. Yogev Y.6:30. American Journal of Obstetrics and Gynecology 2003. Undiagnosed asymptomatic hypoglycemia: diet. Poyhonen-Alho M. Mecacci F. 189. Telemedicine in the treatment of diabetic pregnancy. Reed GF. et al. Dodson S and French L. et al. Langer O. Diabetologia 2002. Sameshima H. 181. and glyburide for gestational diabetic pregnancy. American Journal of Obstetrics and Gynecology 2004. Fotinos C. et al. Do HbA(1)c levels and the self-monitoring of blood glucose levels adequately reflect glycaemic control during pregnancy in women with type 1 diabetes mellitus? Diabetologia 2006. Yogev Y.16(2):138–41. Obstetrics and Gynecology 2004. Moschetta M.2(Suppl 1):S67–71. Chen R.33(6):519–23. 185.15(1):51–5. 182. Kerssen A.35:96A. Journal of Maternal-Fetal and Neonatal Medicine 2004. BMC Pregnancy and Childbirth 2006. 208. 190. Macrosomia despite good glycaemic control in Type I diabetic pregnancy. Diabetes Technology and Therapeutics 2000. Relationship of glycosylated hemoglobin. et al. 203. Mires GJ.53(10):838–41. Jones M and Mandel F. Rodriguez DA. 200. Journal of Perinatal Medicine 2005. American Journal of Perinatology 1994. Parretti E. 202. High spontaneous premature labour rate in insulin-dependent diabetic women: An association with poor glycaemic control. and birthweight macrosomia.

Caprio S. Frost-Larsen K. Grey RH. Larsen HW. Klein BE.18(5):631–7. Miodovnik M. 2004. Temple P. Photocoagulation for diabetic macular edema..26(2):104–12. risks.98(5 Suppl):766–85. Follow-up of the Kroc Collaborative Study. Counterregulatory hormonal responses to hypoglycemia during pregnancy. Critical Care Medicine 2005. Early treatment diabetic retinopathy study report number 1. Santoro S. Acta Diabetologica 2003. Khoury JC. Brimacombe J. Short term parenteral nutrition in a pregnant diabetic woman with hyperemesis gravidarum. Hanker JP and Schneider HP. Hod M. 222. Chew E. et al.23(4):163–5. 227. Conroy C. et al. 226. Ophthalmology 1987.24(12):2078–82. 224. Diamond MP. 220. et al.22(11):1790–5. 206 . and patient satisfaction associated with insulin pump therapy for the pregnancy complicated by type 1 diabetes mellitus. 235. Archives of Ophthalmology 1984. 230. Dalfra MG. 232. Mills JL. Effect of 1 year of near-normal blood glucose levels on retinopathy in insulin-dependent diabetics. Effect of pregnancy on diabetic retinopathy. 249. Shammas A and Haddadin A.94(7):761–74. Carroll MA and Yeomans ER. Metabolic control and progression of retinopathy: The Diabetes in Early Pregnancy Study. Benefits. Early worsening of diabetic retinopathy in the DCCT. Gonzalez C. et al. Dahl-Jorgensen K. 229. Cochrane Database of Systematic Reviews 2007. Diabetes Care 1999. Diabetic retinopathy after two years of intensified insulin treatment. et al. Correlations with regulation of hyperglycemia. Diabetes Care 2001. Temple RC.182(6):1283–91. Moss SE and Klein R. Midazolam and parenteral nutrition in the management of life-threatening hyperemesis gravidarum in a diabetic patient.33(10 Suppl):S347–53. Simmons D. 219. Rapid tightening of blood glucose control leads to transient deterioration of retinopathy in insulin dependent diabetes mellitus: the Oslo study. 241. Kranias G.319(7219):1223–7. Impact of pregnancy on the progression of diabetic retinopathy in Type 1 diabetes. ETDRS report number 9. Acta Medica Scandinavica 1987. American Journal of Obstetrics and Gynecology 1992. Kroc Collaborative Study Group. 248.12(Pt 1):69–76.87(4):568–74.59(6):699–704. et al.18(7):573–7. Sparrow JM. et al. Complications and fetal outcome in diabetic pregnancy. Sampson MJ. Aldridge VA. Sakol P and Metzger BE. Dibble CM. British Medical Journal 1999. Laatikainen L. British Medical Journal 1985.1(8318):200–4. II. JAMA : the journal of the American Medical Association 1986. et al. 225. 245. Diabetic Medicine 2007.(3). 246. Holing E. Diabetic retinopathy in pregnancy during tight metabolic control. et al. Marisavljevic D. Klebe JG and Bek T. et al. et al. et al. Bahrain Medical Bulletin 2001. Lauritzen T. Type 1 diabetes in adults . Masin M. 243. Effect of physician specialty on outcomes in diabetic ketoacidosis. A controlled study of the influence of continuous subcutaneous insulin infusion treatment on diabetic retinopathy during pregnancy. The Wisconsin epidemiologic study of diabetic retinopathy. Treatment techniques and clinical guidelines for photocoagulation of diabetic macular edema. Chew EY. Weiner E. 217. Prevalence and risk of diabetic retinopathy when age at diagnosis is less than 30 years. et al. Levetan CS. Teramo K and Hieta-Heikurainen H. 247. 242. Diabetes Care 1990. Diabetic Medicine 2001. Diabetes Prevention Program Research Group. Miodovnik M. 236. Fink-Cohen S. Diabetes Care 1995. Passaro MD. Twice daily versus four times daily insulin dose regimens for diabetes in pregnancy: randomised controlled trial. Moss SE. Maculopathy.166(4):1214–18. 239. Clinical variables associated with diabetic ketoacidosis during pregnancy.Diabetes in pregnancy 214. Impairment of counterregulatory hormone responses to hypoglycemia in pregnant women with insulin-dependent diabetes mellitus. Archives of Ophthalmology 1986. Expert Opinion on Pharmacotherapy 2005. Zarkovic M. Ophthalmology 1991. 223. Axer-Siegel R. Coustan DR. costs. Kochenour NK and Worley RJ. 231.221(4):367–76. 233. Reece EA. Hanssen KF.23(2):228–30. Lapolla A. Nachum Z. Early Treatment Diabetic Retinopathy Study Research Group. London: Royal College of Physicians.23(8):1084–91. Rosenn BM. Sherwin RS. Rodgers BD and Rodgers DE. Burkart W. Archives of Ophthalmology 1998. Nathan DM. Diabetic retinopathy during pregnancy. Klein R. Effect of pregnancy on diabetic retinopathy. I. JAMA: the journal of the American Medical Association 1988. Archives of Ophthalmology 1985. Metzger BE. Continuous subcutaneous insulin infusion versus multiple daily injections of insulin for pregnant women with diabetes. Effect of pregnancy on microvascular complications in the diabetes control and complications trial the diabetes control and complications trial research group.103(11):1815–19. Acta Obstetricia et Gynecologica Scandinavica 2000. 228. Klein BE.166(1 Pt 1):70–7.103:1796–806. Nesovic M.6(5):735–42. Progression of diabetic retinopathy in pregnancy: Association with hypertension in pregnancy. 216. Diabetic ketoacidosis in pregnancy. Gabbe SG. Early Treatment Diabetic Retinopathy Study Research Group.79(5):367–70. National Collaborating Centre for Chronic Conditions.116:874–86. Tuffnell DJ and West J.36(11):797–800. Early photocoagulation for diabetic retinopathy. Obstetrics and Gynecology 1996.290(6471):811–15. Phelps RL.260(1):37–41.13(1):34–40. Bailey CC. et al.national clinical guideline for diagnosis and management in primary and secondary care. Diabetes Care 2000. Ophthalmology 1996. Brinchmann-Hansen O. Rosenn B. Insulin analogue therapy in pregnancies complicated by diabetes mellitus. Lauszus F. Analysis of outcome of pregnancy in type 1 diabetics treated with insulin pump or conventional insulin therapy. Intensified conventional versus insulin pump therapy. Thompson CF.40(3):143–9.24(2):137–44. 238. American Journal of Obstetrics and Gynecology 1992.104(12):1806–10. Reece EA. Maayah J. American Journal of Obstetrics and Gynecology 2000. Journal of Reproductive Medicine 1991. A randomized clinical trial of the insulin pump vs intensive conventional therapy in diabetic pregnancies. et al. Diabetes Control and Complications Trial (DCCT) Research Group. 234. Lancet 1983.255(5):631–6.102(4):520–6. 240. Archives of Gastroenterohepatology 1995. Effect of pregnancy on progression of diabetic retinopathy. et al. 221. 244. et al. Farrar D. Eye 1998. Gynecologic and Obstetric Investigation 1988. Jablonski KA. et al. 215. 218. Early Treatment Diabetic Retinopathy Study Research Group. Early Treatment Diabetic Retinopathy Study Report Number 2. Changes in diabetes retinopathy during pregnancy. The National Diabetic Retinopathy Laser Treatment Audit. Salzberg S. The prevalence of retinopathy in impaired glucose tolerance and recent-onset diabetes in the diabetes prevention program. 237. Obstetrics and Gynecology 1982. Anaesthesia and Intensive Care 1995.14(1–2):33–5. Ben Shlomo I. Use of insulin pumps in pregnancies complicated by type 2 diabetes and gestational diabetes in a multiethnic community. The Diabetes Control and Complications Trial Research Group. 250. et al.

252. Cincotta RB. 264.105(10):1801–15. Obstetrics and Gynecology 1991. 280. 260. de Leiva A. Comstock CH. Songraphically estimated fetal weights: accuracy and limitations. Italian Journal of Gynaecology and Obstetrics 1995.5(2):114–18. Klein BE. Diabetologia 2001. Severs CP.23(4):426–31. James A. et al.5(5):273–80. McLeod L and Ray JG. Rosenn B.77(4):520–4. Thornton J. Photocoagulation treatment of proliferative diabetic retinopathy: clinical application of Diabetic Retinopathy Study (DRS) findings. Torres CG. Blood pressure and retinopathy in type I diabetes. Albert TJ. et al. Aldington SJ. Matthews DR. Are babies of normal birth weight who fail to reach their growth potential as diagnosed by ultrasound at increased risk? Ultrasound in Obstetrics and Gynecology 1995. Ophthalmology 1981. The stages in diabetic renal disease. Diabetes and Pregnancy. Di Renzo GC.79(1):55–8. et al.35(11):1049–54. 276. Sorensen S and Molsted-Pedersen L. Archives of Ophthalmology 2004. The Wisconsin Epidemiologic Study of Diabetic Retinopathy: XVII. American Journal of Obstetrics and Gynecology 1996. Landon MB. Diabetic Retinopathy Study Research Group.97(2):155–9.318(7175):29]. 259. 258. Kehl RJ. 2003. 283. Harrist RB. Four risk factors for severe visual loss in diabetic retinopathy.19(2):171–6. Scanaill SN. 2002. 486–94. Diabetes Care 2000. Effect of blood pressure control on diabetic microvascular complications in patients with hypertension and type 2 diabetes. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. Stratton IM. Cincotta RB.23(Suppl 2):B54–64. Fearneyhough TC. Archives of Ophthalmology 1979.97(4):654–5. The third report from the Diabetic Retinopathy Study. Acta Obstetricia et Gynecologica Scandinavica 1998. Mid-second trimester fetal echocardiographic examination for detecting cardiac malformations in pregnancies complicated by pregestational diabetes. Wales and Northern Ireland: population based study. Levine AB.44(2):156–63. 261. Ferrero A. Sparrow JM. Journal of Reproductive Medicine 2000. Australian and New Zealand Journal of Obstetrics and Gynaecology 2001. et al. Smith RS. 207 . et al. Fetal growth and body composition in infants of women with diabetes mellitus during pregnancy. Chan FY. 267. Ophthalmology 1990. 262. Feldman K. 257. et al.45(2):117–21.112(11):1461–6. Diabetic vascular complications in pregnancy: nephropathy. et al. Stuart B. EUROCAT Central Registry. et al. Australian and New Zealand Journal of Obstetrics and Gynaecology 2005. 285. 277. Brinchmann-Hansen O. Joner G. III. 281. UK Prospective Diabetes Study Group. 271. Diabetic Medicine 2006. Journal of Maternal-Fetal Medicine 1996. 273. Rosenn BM and Miodovnik M. Reduced prevalence of early preterm delivery in women with Type 1 diabetes and microalbuminuria – Possible effect of early antihypertensive treatment during pregnancy.32(Suppl 2):64–78.7(2):79–82. Giancotti A. Brown B. 282.24(10):804–7.317(7160):703–13. Krew MA. et al. Estacio RO. A nationwide cross-sectional study of retinopathy and microalbuminuria in young Norwegian type 1 (insulin-dependent) diabetic patients. Ultrasound in Obstetrics and Gynecology 2002.41(4):429–32.25(10):927–9. Prenatal Diagnosis 2005.159:118–21. Diabetes Care 2001. Coomarasamy A. Muller PR.13(Pt 2):151–9. [Erratum appears in BMJ 1999.24(10):1739–44. Ophthalmology 1998. Mogensen CE. Feldt-Rasmussen B. et al. Radiology 1985. 256. et al. Maternal diabetes mellitus: which views are essential for fetal echocardiography? Obstetrics and Gynecology 1997. Second-trimester prenatal screening markers for Down syndrome in women with insulin-dependent diabetes mellitus. Garg SK. Spencer K. In: Hod M. Grey RH. 255. 279.154(2):503–5. Bailey JA.122(11):1631–40. Huttly W.77(2):155–8. Moss SE. 274. 270. Muller C. Sonographic diagnosis of the large for gestational age fetus at term: does it make a difference? Obstetrics and Gynecology 1992. 254. Association with cesarean delivery. Damm P. Stratton IM. Routine ultrasound screening in diabetic pregnancies. et al. The influence of maternal insulin-dependent diabetes on fetal nuchal translucency thickness and firsttrimester maternal serum biochemical markers of aneuploidy. Combs CA. p. Wong SF. Marceca M. European Registration of Congenital Anomalities: report 8: surveillance of congenital anomallies in Europe 1980–1999. Thomas A. Sonographic estimation of fetal weight in macrosomic fetuses: diabetic versus non-diabetic pregnancies. Pedersen JF. Macintosh MC. et al. Atzei A. Chan FY. Sehdev HM.174(5):1424–8. Sonographic EFW and macrosomia: is there an optimum formula to predict diabetic fetal macrosomia? Journal of Maternal-Fetal Medicine 2000. et al. Jovanovic L.References 251. The Diabetic Retinopathy Study Research Group. Utility of fetal echocardiogram in high-risk patients. Perinatal mortality and congenital anomalies in babies of women with type 1 or type 2 diabetes in England. Cicero S.90(4 Pt 1):575–9. Gifford N.88(7):583–600. Greene MF and Benacerraf BR. Serum levels of human placental lactogen. Risks of progression of retinopathy and vision loss related to tight blood pressure control in type 2 diabetes mellitus: UKPDS 69. Diabetologia 1992. Hadlock FP. London: Taylor & Francis Group. Newtownabbey: University of Ulster. Kohner EM. British Medical Journal 1998. eds. Ultrasonographic prediction of fetal macrosomia. Use of femur length/abdominal circumference ratio in detecting the macrosomic fetus. et al. Connock M. pregnancy-associated plasma protein A and endometrial secretory protein PP14 in first trimester of diabetic pregnancy. With emphasis on the stage of incipient diabetic nephropathy. Aldington SJ. Damm P. Chase HP. Nielsen LR. Prenatal Diagnosis 2004. BJOG: an international journal of obstetrics and gynaecology 2005. et al. Lockwood CJ. et al. 266. Diabetes 1983. et al. Bancerraf BR. et al.45(1):17–22. Stratton JF. 272. et al. Community Genetics 2002. 269. Ekbom P. et al. Wong SF. 284. 253. et al. Rudnicka A and Wald NJ. 268. Prenatal detection of fetal anomalies in pregnancies complicated by insulin-dependent diabetes mellitus.5(1):33–9. Clinical outcomes. UKPDS 50: risk factors for incidence and progression of retinopathy in Type II diabetes over 6 years from diagnosis. Jackson WE. Lorenz RP. 263. Pregnancy outcome in type 1 diabetic women with microalbuminuria. Wheller JJ. 275. Parry S. Prevention and detection of diabetic embryopathy. Miodovnik M. Accuracy of ultrasound biometry in the prediction of macrosomia: a systematic quantitative review. American Journal of Obstetrics and Gynecology 1988. British Medical Journal 2006. et al. Fleming KM. 278. Klein R. Bailey CC. The National Diabetic Retinopathy Laser Treatment Audit. 265. Langer O. Jeffers BW. DRS report Number 8.333(7560):177. Eye 1999. Christensen CK and Vittinghus E. et al.9(1):55–61. The 14-year incidence and progression of diabetic retinopathy and associated risk factors in type 1 diabetes. Prenatal diagnosis in diabetic gravidas: utility of ultrasound and maternal serum alpha-fetoprotein screening.

Wong SF. betahydroxybutyrate. American Journal of Perinatology 1991. Mordes D. Jr. The influence of obesity and diabetes on the risk of cesarean delivery. Durnwald CP. Use of umbilical artery Doppler velocimetry in the monitoring of pregnancy in women with preexisting diabetes. Effects of acetylsalicylic acid on plasma glucose. 305. insulin and C-peptide. Can ultrasound fetal biometry predict fetal hyperinsulinaemia at delivery in pregnancy complicated by maternal diabetes? European Journal of Obstetrics. et al. 316. British Medical Journal 1993. Sermer M.34(2):353–67. Obstetrics and Gynecology 1994. and Reproductive Biology 1992. The effect of ritodrine on carbohydrate and lipid metabolism in normal and diabetic pregnant women. Geslevich Y. 315. Naylor CD.14(4):235–42.99(2):135–40. Reece EA. Insulin dose during glucocorticoid treatment for fetal lung maturation in diabetic pregnancy: test of an algorithm [correction of an algorithm]. 291. 308. Doppler ultrasound for fetal assessment in high risk pregnancies. European Journal of Obstetrics. Kolderup LB. Chan FY. Incidence of persistent birth injury in macrosomic infants: association with mode of delivery. 319.15(6):437–40. Mathiesen ER. Gibson J and Railton A. Kuhl C. Salvesen K.32(10):798–800. Ola B. 288. Acta Endocrinologica 1979. Kernaghan D. Is there a correlation between aortic Doppler velocimetric findings in diabetic pregnant women and fetal outcome? Journal of Ultrasound in Medicine 1996.248(8):973–5. Schilthuis MS and Aarnoudse JG. A protocol for improved glycaemic control following corticosteroid therapy in diabetic pregnancies. Bricker L and Neilson JP. et al. free fatty acid. Figueroa R. Journal of Ultrasound in Medicine 1996. 294. Steel JM. et al. (Cochrane Review). Lam H. and Reproductive Biology 2007.83(3):462–5. Laros RK. Penry M and Swain M. Assimakopoulos E.Diabetes in pregnancy 286.275(15):1165–70.92(4):669–79.24(5):534–7. Diabetes mellitus in pregnancy and the assessment of umbilical artery waveforms using pulsed Doppler ultrasonography. et al. 314. 320. et al. Flyvbjerg A. Richards SR and Klingelberger CE. glucagon and C-peptide responses to salbutamol in insulin-dependent diabetic subjects.163(2):509–10.124(2):144–9. 306. American Journal of Obstetrics and Gynecology 1990. Chichester: Wiley Interscience. A case report. diabetic women without preeclampsia: The role of metabolic control. Acta Obstetricia et Gynecologica Scandinavica 2002.191(3):969–74.43(4):302–6. Haenel LC. 312.8(4):273–7. (Cochrane Review). Doppler velocimetry of the umbilical artery in pregnancies complicated by insulin-dependent diabetes mellitus. 290. Moroder W. New Zealand Medical Journal 2006. American Journal of Obstetrics and Gynecology 2003. Gynecology. Farrell T. American Journal of Obstetrics and Gynecology 2004. Journal of Reproductive Medicine 1987. Ultrasound in Obstetrics and Gynecology 2004. Diabetes Care 2006. RCOG Press. 293. Persson B. JAMA: the journal of the American Medical Association 1982. Catalano P. 309. Lunell NO. et al. In: Cochrane Database of Systematic Reviews. Ciba Foundation symposium 1978. Examination of factors contributing to the risk of cesarean delivery in women with gestational diabetes. and obstetrical complications in women with pregestational diabetes: a population-based study in Ontario. In: Cochrane Database of Systematic Reviews. Torella R. Comparison of umbilical Doppler velocimetry. 311. Zimmermann P. et al. Screening for fetal well-being in a high-risk pregnant population comparing the nonstress test with umbilical artery Doppler velocimetry: a randomized controlled clinical trial. Journal of Ultrasound in Medicine 1994. et al. 313. Diabetic ketoacidosis and insulin resistance with subcutaneous terbutaline infusion: a case report. Giugliano D.81(9):835–9. et al. Tibaldi JM. Fetal death associated with severe ritodrine induced ketoacidosis. Owen P. Passariello N.13(2):73–80. Halpren EW. nonstress testing. 2000. Royal College of Obstetricians and Gynaecologists. American Journal of Obstetrics and Gynecology 1997. et al. Hagay Z. Wang P. Cincotta RB. et al.29(2):232–5. Lenz S. Hagay Z. Analyses of the ALSPAC cohort of 12 944 women in England. et al. Routine ultrasound scanning in pregnancy. 2001. et al.307(6911):1064. Clinical and ultrasound prediction of macrosomia in diabetic pregnancy. Lancet 1980. Fredholm BB. Gynecology. British Journal of Obstetrics and Gynaecology 1996. et al.15(4):301–8. Fraser RB. 304. Fuglsang J. London. Razzaq A. et al. Fetal size and growth velocity in the prediction of the large for gestational age (LGA) infant in a glucose impaired population. Naeye RL. Australian and New Zealand Journal of Obstetrics and Gynaecology 2003. Doppler umbilical artery flow velocity waveforms in diabetic pregnancy. Battino S. carbohydrate and lipid metabolites in diabetic and non-diabetic women. Intravenous ritodrine as a possibly provocative predictive test in gestational diabetes. Diabetic Medicine 2003. European Journal of Obstetrics and Gynecology 2007. Chen E. Journal of the American Osteopathic Association 1988.188(5):1366–71.(63)227–41. Bebbington M. 300. 307. Leung WC. Metzger W.20(1):73–5.132(2):189–92.18(1):27–36. 321. Reece EA. Doppler study of the umbilical and fetal middle cerebral arteries in women with gestational diabetes mellitus. 303. et al. and biophysical profile in pregnancies complicated by diabetes. Steel JM. Chichester: Wiley Interscience. Hellmuth E. and Musci TJ. Preterm delivery in normoalbuminuric. Bernstein IM and Catalano PM. et al. Maharaj D. The outcome of diabetic pregnancies: a prospective study. Lorber DL and Nerenberg A. Trends in deliveries. Williams KP. Gynecology. Feig DS. et al. Kaushal K. Byrne DW. Neilson JP and Alfirevic Z.119(1241):U2146. Reliability of ultrasound estimation of fetal weight in term singleton pregnancies. 297. Kreutner K. Ehrenberg HM. 310. 289. Bracero LA. Cesarean delivery in relation to birth weight and gestational glucose tolerance. 299. Dangers of intravenous ritodrine in diabetic patients. Christensen AB. 208 . et al. British Journal of Obstetrics and Gynaecology 1992. Tocolytic drugs for women in preterm labour. 301. Johnstone FD. 296. Kernaghan D. A random single Doppler study of the umbilical artery in the evaluation of pregnancies complicated by diabetes. 317. Lauszus FF. 292. and Reproductive Biology 2006. Haddad NG.103(8):747–54. 2002. American Journal of Perinatology 1995. Colman A. Lee CP. Kofinas AD. Kujansuu E and Tuimala R.88(2):241–4. Actions of salbutamol in late pregnancy: plasma cyclic AMP. 302. Johnstone FD. Patel RR. European Journal of Obstetrics.177(1):37–41. Farquharson DF. Peters TJ and Murphy DJ. Sykora K.131(2):146–50. Canada. Pathophysiology or practice style? JAMA: the journal of the American Medical Association 1996. Acta Diabetologica Latina 1981. Prescott RJ. prenatal care.12(6):437–8. 318. Ben-Ami M. 298. 1996–2001. Diabetologia 1978.1(8178):1145. et al.47(2):85–93. Prenatal risk factors for Caesarean section. 295. et al. Routine doppler ultrasound in pregnancy. Ketoacidosis secondary to oral ritodrine use in a gestational diabetic patient: Report of a case. 287. et al. Uteroplacental Doppler flow velocity waveform analysis correlates poorly with glycemic control in diabetic pregnant women. Hutton J. International Journal of Epidemiology 2005. Soifer NE.

et al. 340. Hypoglycemia in infants of diabetic mothers: experience in a rural hospital. et al. American Journal of Obstetrics and Gynecology 2004. fetal and immediate neonatal morbidity and operative delivery.73(2):105–11. Diet-controlled gestational diabetes mellitus does not influence the success rates for vaginal birth after cesarean delivery.11(Suppl 1):63–72. Hebl JR.27(3):623–41. 350. Lattermann R. Taylor R.16(7):573–8. Medical concerns in the neonatal period. Fassett MJ. Kjos SL. Adelantado JM. 352.43(1):1–6. Insler V and Hagay ZJ.400A.169(3):611–15. Obesity and obstetric anaesthesia.185(4):916–19.23(2):105–10. 330. Obstetrics and Gynecology 1997.184(6):1104–7. Acta Obstetricia et Gynecologica Scandinavica 1988. Datta S. Graves W. et al. 324. Tsen LC. 344. Randall H. 323. 345. Levy AL. Carron Brown S. Rosenberg VA.89(6):913–17. Carli F.195(4):1095–9. 325. et al. Acta Diabetologica 1994. 348. et al. Gonzalez JL. 327. Anesthesiology 2002. 335. Induction of labor at 38 to 39 weeks of gestation reduces the incidence of shoulder dystocia in gestational diabetic patients class A2. Wykes L. Clinics in Obstetrics and Gynaecology 1982. A comparative study of constant intravenous insulin infusion and continuous subcutaneous insulin infusion pump (CSIIP). Lee C. Gilson GJ. Intrapartum maternal glycemic control in women with insulin requiring diabetes: a randomized clinical trial of rotating fluids versus insulin drip.47(11):931–2. Perinatal asphyxia in infants of diabetic mothers is associated with maternal vasclopathy and hyperglycaemia in labour. 351. Tsai M and Tyson JE. Clinics in Perinatology 2000.178(5):922–5. Anesthesia and Analgesia 1991. Diabetic Medicine 1999. Insulin management during labour and delivery in mothers with diabetes. Hod M. Influences on neonatal outcome.103(5):1294–9. et al. Saravanakumar K. et al. 355. et al.References 322. Holt VL and Mueller BA. 354. Acid-base status of diabetic mothers and their infants following spinal anesthesia for cesarean section.99(4):537–41. Paediatric and Perinatal Epidemiology 1997. Diabetes Care 1998. 349. 346. Elkousy M. 341. Gonen O. Peled Y. Nutrition and Metabolism . Clinical Obstetrics and Gynecology 2003. et al. Slocum J. Acta Paediatrica Scandinavica 1985. Dicker D. Kopp SL.17(2):113–15. Kyne-Grzebalski D. American Journal of Perinatology 1987. Neonatal Epidemiology & Followup 1987. Emergencies and common abnormalities involving the skin. Gestational diabetes mellitus: Metabolic control during labour. head. Izquierdo LA. American Journal of Perinatology 2006. Influence of the maternal plasma glucose concentration at delivery on the risk of hypoglycaemia in infants of insulin-dependent diabetic mothers. Clinical outcomes of pregnancy in women with type 1 diabetes. Rosen DJ. 343. Intrapartum management of insulin-dependent diabetes mellitus (IDDM) gestants. 338. Kjos SL. et al. Neurologic complications after neuraxial anesthesia or analgesia in patients with preexisting peripheral sensorimotor neuropathy or diabetic polyneuropathy. Conway DL and Langer O. Aucott SW. Evaluation and treatment of thrombosis in the neonatal intensive care unit. Kyne-Grzebalski D.Clinical and Experimental 2000. Anesthetic management of the parturient with cardiac and diabetic diseases. et al. Anaesthesia 2006. Rappaport VJ. Timing and mode of delivery in gestational diabetes. 353. Perioperative maternal and neonatal acid-base status and glucose metabolism in patients with insulin-dependent diabetes mellitus. Pearson DW and Sutherland HW. Rees GA.31(3):126–9.5(2):265–92. Journal of Perinatology 1997.190(3):790–6. Maternal. Williams TG. et al. Diabetic Medicine 1990. pregnancy and anaesthesia. et al. Hertz RH. 357. Mimouni F and Tsang RC.400A. American Journal of Obstetrics and Gynecology 2001.7(2):162–4. Mimouni F. American Family Physician 2002. Dilling LA and Vidyasagar D. American Journal of Obstetrics and Gynecology 2001. Vaginally administered misoprostol for outpatient cervical ripening in pregnancies complicated by diabetes mellitus. Induction of labor versus expectant management in macrosomia: a randomized study. Incerpi MH. Khonjandi M. Gestational diabetes: the dilemma of delivery.74(2):268–73. 209 . Ramanathan S. 337. 332. Vaginal birth after cesarean among women with gestational diabetes. Weitzen S. et al. American Journal of Obstetrics and Gynecology 1998. Khoo P and Arismendy J. Obstetrics and Gynecology 1974. Risk of cesarean wound complications in diabetic gestations. neck. Bar J. et al. Obstetrics and Gynecology 2002. Anesthesia and Analgesia 2006. Naulty JS. Elective delivery of infants with macrosomia in diabetic women: reduced shoulder dystocia versus increased cesarean deliveries. Effect of management policy upon 120 Type 1 diabetic pregnancies: policy decisions in practice. et al. Epidural blockade modifies perioperative glucose production without affecting protein catabolism. Journal of Reproductive Medicine 2002. Antepartum management protocol. Akera C and Ro S. Modanlou H and Dorchester W. Schroeder DR. Insulin-requiring diabetes in pregnancy: a randomized trial of active induction of labor and expectant management. 347. Eglinton GS. Montoro M.9(2):311–32. Effect of labor induction on cesarean section rates in diabetic pregnancies. 334. American Journal of Obstetrics and Gynecology 2006. et al. Rao SG and Cooper GM. Lean ME. 356.191(3):958–63. Diabetes. Kitzmiller JL. Henry OA.65(1):61–8. 329. Stevens E. check. Coleman TL. Hayes TM and Pearson JF. American Journal of Perinatology 1996.67(4):333–8. Marchiano D. 333. Feldberg D. Rigorous management of insulin-dependent diabetes mellitus during pregnancy. Mwangi B.82(1):94–7.46(3):700–10. Dolfin Z. Clinics in Family Practice 2003. Miodovnik M. Relative effects of antepartum and intrapartum maternal blood glucose levels on incidence of neonatal hypoglycemia. 342.21(Suppl 2):B113–17.13(5):257–62. 339. Neonatal Epidemiology & Follow-up 1987. et al.61(8):662–5. Hypoglycemia in infants of diabetic mothers: effect of epinephrine therapy. and respiratory and cardiovascular systems. Attempt and success rates for vaginal birth after caesarean section in relation to complications of the previous pregnancy. American Journal of Obstetrics and Gynecology 1993. The Journal of Pediatrics 1973.61(1):36–48. Balsells M. et al. 336. Van Howe RS and Storms MR. Diabetes. The newborn examination: Part I. Lurie S. Schmolker L. American Journal of Obstetrics and Gynecology 2004. 331.97(2):374–81. Samuel N. Rauch ER.4(2):106–14. Haworth JC. Curet LB. Management of the insulin-dependent diabetic during labor and delivery. Corcoy R. Fuloria M and Kreiter S. Andersen O. Hertel J. 326. Takoudes TC.13(5):293–6. 328. Anesthesia and Analgesia 1982. Edstrom CS and Christensen RD.

High amniotic fluid erythropoietin levels are associated with an increased frequency of fetal and neonatal morbidity in type 1 diabetic pregnancies. Steindel E. Harder T. Neonatal Network . Neale L. Obesso A. 210 . Transfer of glyburide and glipizide into breast milk.20(3):340–2. Briggs GG. Transient neonatal hypoglycemia--long-term effects on neurodevelopmental outcome.75(3):261–90. Health Care for Women International 1992. Impact of breast-feeding on psychomotor and neuropsychological development in children of diabetic mothers: role of the late neonatal period. Nageotte MP. Cordero L. Scottish Intercollegiate Guidelines Network. Rennie JM. 394.28(8):1851–5. Patterns of metabolic adaptation for preterm and term infants in the first neonatal week. et al. Rodekamp E. Breastfeeding Review 2006.68(3):329–43. Diabetes Care 1994. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Milk of patients with tightly controlled insulin-dependent diabetes mellitus has normal macronutrient and fatty acid composition. Journal of the College of Physicians and Surgeons – Pakistan: JCPSP 2006. et al.17(1):13–19. 390. 379.20(6):17–23. Cow’s milk exposure and type I diabetes mellitus. IDDM and early infant feeding. et al. Franke K. 376. 384. 364. Adverse neurodevelopmental outcome of moderate neonatal hypoglycaemia. Ingardia CM. 365.105(5):1141–5. Journal of the American Dietetic Association 1988. 393. Treuer SH. Journal of Pediatric Endocrinology 2002. The breast-feeding experience of women with type I diabetes. 380. Boyes S. Ferris AM. Hildebrandt R. Clark JD. Diabetes Care 2002. Feig DS. Marasigan BV. 378. et al. Diabetes Care 1997. Kohlhoff R. Soysal S. 363. 389. et al. [Erratum appears in J Coll Physicians Surg Pak 2006. Lucas A.92(1):32–6. Diabetes Care 2005. Hawdon JM. Ward Platt MP and Aynsley-Green A. et al. Meloni T. Morley R and Cole TJ. Expressing and storing colostrum antenatally for use in the newborn periods. Immediate or later feeding for premature babies? A controlled trial. Kari MA.297(6659):1304–8. 366. Rodekamp E. Ergenekon E. Mannazzu MC. Lucas A. Newton KM and Knopp RH. Harder T. Leff EW and Jefferis SC. Diabetologia 2004. Belfast: CREST. Pediatrics 2000. Management of diabetes. Diabetes Care 2004. Archives of Disease in Childhood 1981.298(6684):1357–8. Williams AF. Diabetes Care 2005.88(3):317–22.16(8):566 Note: Akhtar. et al.43(6):429–32. Gagne MP. Gerstein HC. et al. Hawdon JM. 2001.152(3):249–54. Obstetrics and Gynecology 2005. Bloom SR.105(6):1437–41.55(3):792–7. 370. Management of Diabetes in Pregnancy. 395. Jensen DM. Impact of early neonatal breast-feeding on psychomotor and neuropsychological development in children of diabetic mothers. Rowan J. Archives of Pediatrics and Adolescent Medicine 1998. ASM]. Lancet 1965. 373. Wharton BA and Bower BD. Knopff A. Bulletin of the World Health Organization 1997. et al. et al. et al. Gaspar R. Teramo K.25(10):1862–8. Dalgic N.15(3):319–24. Plagemann A. 368.25(1):16–22. Effect of fluid and calorie deprivation on blood glucose.14(3):16. Neonatal complications in infants born to diabetic mothers. Admissions to neonatal intensive care unit following pregnancies complicated by gestational or type 2 diabetes. et al. Excretion of metformin into breast milk and the effect on nursing infants. Halliday HL. et al.34(6):490–6. II. Metabolic and endocrine responses to a milk feed in six-day-old term infants: differences between breast and cow’s milk formula feeding. 375. Paediatrics and Child Health 2004. Raza SJ. 362. Alam M. Journal of Perinatal Medicine 2006. Acta Paediatrica Scandinavica 1981.7(1):87–92. Beard AG. et al.Journal of Neonatal Nursing 2001. Gestational diabetes and its impact on the neonate.70(2):195–200. et al. Long-term impact of breast-feeding on body weight and glucose tolerance in children of diabetic mothers: role of the late neonatal period and early infancy. 382. British Medical Journal 1989. 387. SM [corrected to Akhtar. Baumgarten A. et al. et al. Bendel RB. Lactation outcome in insulin-dependent diabetic women.58(1):43–8. Diabetes Care 2002. Perinatal lactation protocol and outcome in mothers with and without insulin-dependent diabetes mellitus. Increasing incidence of diabetes after gestational diabetes: a long-term follow-up in a Danish population. A meta-analysis of infant diet and insulin-dependent diabetes mellitus: do biases play a role? Epidemiology 1996. Norris JM and Scott FW. Hansen T. American Journal of Clinical Nutrition 1993. Dalidowitz CK. 391. Australian and New Zealand Journal of Obstetrics and Gynaecology 2003. Long-term impact of neonatal breast-feeding on body weight and glucose tolerance in children of diabetic mothers.13(3):249–60. 1–74. British Medical Journal 1988. van Beusekom CM. Neubauer SH. Is there a relationship between metabolic control and glucose concentration in breast milk of type 1 (insulin-dependent) diabetic mothers? Experimental and Clinical Endocrinology 1988. Hypertrophic cardiomyopathy in infants of poorly-controlled diabetic mothers. The Journal of Pediatrics 1966. 360. A national clinical guideline. et al. Kohlhoff R. Diabetes 2006. Practical Diabetes International 2003. 377.20(8):271–5.Diabetes in pregnancy 358. Cornblath M.57(6):938–43. Zeegers TA. Hypoglycaemia of the newborn: a review. Watson D. et al. Jones CW. Cox SG. Predictors of postpartum diabetes in women with gestational diabetes mellitus.47(10):1695–703. Harder T. American Journal of Clinical Nutrition 1993.67(4 Spec No):357–65. et al. 361. CREST. Plagemann A.9(10):723–40. 2001. Screening guidelines for newborns at risk for low blood glucose. Panos TC. Insulin requirements of diabetic women who breast feed.28(6):1457–62. Ambrose PJ. 369. 374. Edinburgh: SIGN. Management of infants of diabetic mothers.16(3):212–15. Davies HA. Landon MB. Dalton KJ. Eronen M. Sherali AR. Martini IA. Kim C. Marinaro AM.28(3):573–8. Lobner K. 392. Ferris AM. 388. 383. 386.2(7420):769–72. 371. et al. Glycaemic behaviour during lactation: postpartum practical guidelines for women with type 1 diabetes. et al. Saez-de-Ibarra L. Edinburgh: Churchill Livingstone. Robertson’s Textbook of Neonatology. Lauenborg J. 396. p. Archives of Disease in Childhood 1992. Controversies regarding definition of neonatal hypoglycemia: suggested operational thresholds. Harder T. Ratzmann KP. Williams AF.27(5):1194–9. 2005. 372. Kraemer JM. 381. et al. Perinatal stress and the premature neonate. Diabetes Care 2005. Kohlhoff R. Briggs GG. 367. A critical overview of the clinical literature. Sardinian case–control study. 359.56(4):258–63.

Gauvreau K. Natural course of gestational diabetes mellitus: Long term follow up of women in the SPAWN study. Requena A. et al. Cardiovascular malformations in infants of diabetic mothers. American Journal of Cardiology 1999. American Journal of Obstetrics and Gynecology 2003. et al. Meriggi E. insulin resistance. 432. 422. et al. Is postnatal oral glucose tolerance testing necessary in all women with gestational diabetes. Diabetes Research and Clinical Practice 2004. Sailer S. 425. et al. Ronneby H and Damm P. Most O. Lindstrom J. Lifestyle change after gestational diabetes.19(4):89–95. Diabetes Care 2006. et al. Cost-benefit analysis of preconception care for women with established diabetes mellitus.405–6. Visser GHA. Halaska M. 2003 [www. Diabète & Métabolisme 1986. Jonsson EK. Stamilio DM. Aberg AE. Australian and New Zealand Journal of Obstetrics and Gynaecology 1997. Diabetes Care 2000. Drexel H. Prevention of perinatal morbidity by tight metabolic control in gestational diabetes mellitus. Predictive factors of developing diabetes mellitus in women with gestational diabetes. Patterns of congenital anomalies and relationship to initial maternal fasting glucose levels in pregnancies complicated by type 2 and gestational diabetes. weight. Yogev Y. Insulin-sensitizing agents: Use in pregnancy and as therapy in polycystic ovary syndrome. 429.12(3):121–9. 401. Singapore Medical Journal 1996. Jarvela IY. British Medical Journal 2007. 412. 428. 424. Pettitt DJ. 414. Bauman AE. Checa MA.334(7588):299–302.14(9):762–5. et al. Sanderson D. Bagust A. 419. Yeo SH and Liauw PC.20(7):594–8. et al. Women’s responses to two models of antepartum high-risk care: Day stay and hospital stay. Increased fetal adiposity: a very sensitive marker of abnormal in utero development. Prevalence and predictors of diabetes mellitus after delivery. 423. Holt RI. 416. et al. Prenatal diagnosis of structural heart disease: does it make a difference to survival? Heart 2002. Long-term prognosis of women with abnormal glucose tolerance in pregnancy. Martan A. Fethney J.22(S1):66–7. Cost-effectiveness of flexible intensive insulin management to enable dietary freedom in people with Type 1 diabetes in the UK. Eskilsson I. Beischer NA and Sheedy MT. 409.34(5):507–10. Sullivan ID. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. Stage E. Goldenberg N. Are diabetic pre-pregnancy clinics worth while? Lancet 1992. Ritchie K. Glueck CJ. Tolerance and effectiveness of propiverine hydrochloride in 752 patients with symptoms of hyperactivity of the detruser. Wang Y. Langer O. Gestational diabetes: the consequences of not treating. Knowler WC. Odibo AO. Boynton J.63(1):67–72. Kitzmiller JL. Continuous subcutaneous insulin infusion versus intensive conventional insulin therapy in type I and type II diabetic pregnancy. et al. et al.23(4):460–4. Abrams KR. 420. et al.89(10):1217–20. Diabetic Medicine 1997. Catalano PM. et al. et al. Shearer A. et al. Wang P. Cheung NW. Incidence of diabetes mellitus in women following impaired glucose tolerance in pregnancy is lower than following impaired glucose tolerance in the non-pregnant state. Koskela P.28(6):779–84. 406. 413. et al. American Journal of Obstetrics and Gynecology 2000. Tan YY. Barrett-Connor E. Diabetes 2006. Human Reproduction Update 2005. Postpartum physical activity and related psychosocial factors among women with recent gestational diabetes mellitus. 404. 426. Studies of postnatal diabetes mellitus in women who had gestational diabetes. et al. et al. 418. Clarke P. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. Huston-Presley L. Trossarelli GF. et al.11(10):761–8. Diabetic Medicine 2003. Goddard JR. Safety and perinatal outcome in pregnancy: a randomized trial comparing insulin aspart with human insulin in 322 subjects with type 1 diabetes. Ogge G. Circulation 1999.83(12):1649–53. Should all pregnant diabetic women undergo a fetal echocardiography? A cost-effectiveness analysis comparing four screening strategies.21(5):460–7. 405. et al. Coassolo KM. Prenatal Diagnosis 2006. Storlien LH. New England Journal of Medicine 2001. American Journal of Obstetrics and Gynecology 2005.344(18):1343–50. 415. Ultrasound in Obstetrics and Gynecology 2006.346(6):393–403.109(11):1227–31. 211 . Gestational diabetes identifies women at risk for permanent type 1 and type 2 diabetes in fertile age: predictive role of autoantibodies. et al. testosterone and development of gestational diabetes: Prospective longitudinal assessment of women with polycystic ovary syndrome from preconception throughout pregnancy. Buchanan TA. Lambert PC. 431.26(1):39–44. Weschler JM. 407.nhshealthquality.62(5):259–64.28(11):2650–4.References 397. Birrell G and Hawthorne G.340(8820):656–8. Hanson RL. Metformin during pregnancy reduces insulin. Kumar RK. Bonnet D. Newburger JW. 398. Wein P. Linne Y. Diabetes Care 2005. Wren C. Jenkins AB. et al. Dietary variables and glucose tolerance in pregnancy. Australian and New Zealand Journal of Obstetrics and Gynaecology 1994. 403. Maccanti S. Simmons D and Robertson S. 408.pdf]. Coltri A. 421. Diabetologia 1996. Thomas A.99(7):916–18. insulin secretion.87(5). Carta Q. Smith BJ. Gregory R and Tattersall RB. BJOG: an international journal of obstetrics and gynaecology 2002.37(4):384–8. A postnatal fasting plasma glucose is useful in determining which women with gestational diabetes should undergo a postnatal oral glucose tolerance test. Ceska Gynekologie 1997.19(3):510–21. Consultation report.182(2):313–20. Diabetes Care 1993. Butera G. 399. Women and Birth 2006. et al. Elixhauser A. Xiang A. Acta Obstetricia et Gynecologica Scandinavica 2002. Wong HS. et al. Barkeling B and Rossner S. 427. Gillies CL. Bradbury I. Salvador C.39(11):1334–7. Fowler SE. Eriksson JG. Diabetes Care 1988. Human Reproduction 2004. et al.192(4):989–97. Juutinen J. Detection of transposition of the great arteries in fetuses reduces neonatal morbidity and mortality. 400. et al.11(4):375–90.16(8):1146–57.55(Suppl 1):A 417. Part 2. Tuomilehto J. Damm P. Comparison of outcome when hypoplastic left heart syndrome and transposition of the great arteries are diagnosed prenatally versus when diagnosis of these two conditions is made only postnatally. 417. New England Journal of Medicine 2002. Schaefer-Graf UM. Influence of maternal insulin treatment on the infants of women with gestational diabetes. Lohan M. Lee CP. 430.189(6):1698–704. 410. Stainton MC. Narayan KM.81(1):11–16. increased sensitivity and irritability of the urinary bladder: Results of an investigation of the use of a drug. Diabetic Medicine 2004.org/nhsqis/files/Ultrasound%20CAR. NHS Quality Improvement. Jouannic JM and Fermont L. 411. Pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people with impaired glucose tolerance: systematic review and meta-analysis. Bichler A. et al.29(3):607–12. Ultrasound in Obstetrics and Gynecology 2003.37(4):420–3. 402. Voigt R. Heart 2003. Chan FY. Hod M. Impact of prenatal diagnosis on perinatal care of transposition of the great arteries. Bonnet D. Gaglioti P. Routine Ultrasound Scanning Before 24 Weeks of Pregnancy. et al. Prenatal screening for congenital heart disease with four-chamber and outflow-tract views: A multicenter study.

Yu D. Zawacki CM. Barden TP and Miodovnik M. 452. Abramowicz JS. The Costs of Induction of Labour by Prostaglandin E2 or Oxytocin: Refining the Estimates.12(5):342–8. Touzet S. Guangzhou. Weeks JW. Clinics in Perinatology 1993. Lepore G.103(2):122–9. Iqbal S.22:418–21. Schmidt MI.11(Suppl 1):s28–s33. Williams CB. 2006. Unit Costs of Health and Social Care. 445. Branchtein L. The experience of diagnosis the abnormal fetal heart by fetal echocardiography to 900 fetuses.83(1):46–51. Ward MK. Lavin JP. Fox HE et al. 434. Effect of varying threshold and selective versus universal strategies on the cost in gestational diabetes mellitus. Brazilian Study of Gestational Diabetes (EBDG) Working Group. Canterbury: Personal and Social Services Research Unit University of Kent at Canterbury. 453. 2006. Parker D. Pan W. 462. Rana S and Abramowicz S.195(2):601–6. York: University of York. Fertility: Assessment and Management for People with Fertility Problems.99(5 Pt 1):740–4. Casadidio I et al. Failure of intensive fetal monitoring and ultrasound in reducing the stillbirth rate. 447. Larijani B. 454.28(6):1482–4. Balaji MS. Nelson R. Diabetes Care 1999. 456. undated [available from www. Reichelt AJ. China: Guangdong Cardiovascular Institute. Sculpher M. Joint Formulary Committee. Wu GP. Acta Diabetologica 2002. 451. 2004. Puerto Rico Health Sciences Journal 2002. Obstetrics and Gynecology 2002. Screening for gestational diabetes mellitus: a decision and cost-effectiveness analysis of four screening strategies.Clinical and Experimental 1993. et al. 450. Acta Obstetricia et Gynecologica Scandinavica 2004. Assessment of glomerular filtration rate during pregnancy using the MDRD formula. et al. Cost-effectiveness analysis of gestational diabetes mellitus screening in France. Chauhan SP. Nutrition and Metabolism . Screening for gestational diabetes in a high-risk population. Fleisher LA. 2004. Franco LJ. Rocher L. Diabetes Care 2005. American Journal of Obstetrics and Gynecology 2006. et al. Spichler ER. Clinical experience with a screening program for gestational diabetes. Brown MB. Major CA.21(2):123–5. Davies L and Drummond M. Coustan DR. et al.19(6):751–5. Danilenko-Dixon D. Decision Modelling for Health Economic Evaluation. London: British Medical Association and Royal Pharmaceutical Society of Great Britain. 457. Goetzl L and Wilkins I. Reed BD. Li YF. Haberman S and Byrne DW.22:418–21. Gorab R et al. Balaji V. 459. Gestational diabetes mellitus in India. Oxford: Oxford University Press. Nicholson WK. Effect of selective screening for gestational diabetes.115(1):109–12. Diabetes Care 1999. de Veciana M et al. Maternal glycemic control and umbilical artery Doppler velocimetry. Journal of Family Practice 1984. 460. Sonographic estimate of birth weight among high-risk patients: feasibility and factors influencing accuracy. Zawacki CM. O’Connor PJ. Hossein-nezhad A and Vassigh A-R.unepsa. Fetal cheek-to-cheek diameter in the prediction of mode of delivery.Diabetes in pregnancy 433. 435. 444. British National Formulary. London: RCOG Press. American Journal of Obstetrics and Gynecology 1999. 449. Journal of Family Practice 1987. 461. Nosari I. et al. Briggs A. Best G and Pressman EK. Repeated random blood glucose measurements as universal screening test for gestational diabetes mellitus.20(3):593–602. 441.181:798–802. BJOG: an International Journal of Obstetrics and Gynaecology 2008. Van Winter J. Maglio ML. Bracero LA. de la Vega A and Verdiales M. Effect of selective screening for gestational diabetes. 436. Ostlund I and Hanson U. 458. Ogburn P. Journal of Obstetrics and Gynaecology 1991. National Collaborating Centre for Women’s and Children’s Health. 52nd ed. Di CG. 440. Diabetes Care 1998. Gestational diabetes: Does the presence of risk factors influence perinatal outcome? American Journal of Obstetrics and Gynecology 1994. Claxton K.52:707–11.39(2):69–73. Methods of screening for and diagnosing of gestational diabetes. Seshiah V. 455. Davies L and Drummond M. et al. 448. 443. Ultrasonographic prediction of birth weight in diabetic pregnancies. Journal of the Association of Physicians of India. Screening for gestational diabetes--analysis by screening criteria. Moran P. Volpe L. Fasting plasma glucose is a useful test for the detection of gestational diabetes. Journal of Perinatology 2002. Curtis L. 171:1003–7. European Journal of Obstetrics.25(6):569–75. Netten A. et al. Journal of Maternal-Fetal and Neonatal Medicine 2002. 437.22(5):403–6. American Journal of Obstetrics and Gynecology 2005. 442.6(1):33–7.HTM. Shields D. Is continuous subcutaneous insulin infusion more effective than intensive conventional insulin therapy in the treatment of pregnant diabetic women? Diabetes.org/china/ab/1327. 439. 1993. 438. Nucci LB. Universal versus selective gestational diabetes screening: application of 1997 American Diabetes Association recommendations. 212 . 446. Smith MC. Archives of Iranian Medicine 2004. Herman WH. Poncet B. American Journal of Obstetrics and Gynecology 1981. 2006. Williams CB. Glyburide compared to insulin for the treatment of gestational diabetes mellitus: A cost analysis.21:1246–9. Iqbal S. Universal screening and intensive metabolic management of gestational diabetes: cost-effectiveness in Italy. Massion C.192(4):1205–11.141(5):491–4.7(4):267–71. Gynecology and Reproductive Biology 2002. accessed 30 August 2006]. Management of labour: consumer choice and cost implications.

156 glycaemic control. measurement. 50 discontinuation in pregnancy. 144 discontinuation in pregnancy. 18 recommendations. 86 planned vs unplanned pregnancies. 22. 54 age (maternal). see insulin (endogenous) volume. 22. 124 glycaemic control and. 108. 53 evidence statement. 111 32 weeks. 62. 108 NICE guideline. see also preconception care. 108. with/without preconception care. epidural/spinal anaesthesia. 114 preterm birth. 119. 71 self-management vs insulin initiation based on. 107 existing guidance. 125 neuropathy effects. oral antihypertensive agents. 23 clinical questions. 107 key recommendations. 180 acid–base status. 11. see obesity drugs aortic velocity waveform. 14 research recommendations. severe. 114 non-administration. 124 importance of full medical history. miscarriage risk and. gestational diabetes risk. 109. risks. 107 first appointment. 110 31 weeks. see general anaesthesia. 11. see ACHOIS study best available evidence. 50 evidence statement. 111 41 weeks. 98. xii asphyxia. 166 intervention and care included. 175 albumin. 72 safety/low risk to fetus. 58. 124 benefits and risks. 114 antenatal.Index abdominal circumference (fetal). see angiotensin-receptor blockers (ARBs) angiotensin-receptor blockers (ARBs) avoidance in pregnancy. 49 adverse pregnancy outcome. 108 from evidence to recommendations. 82. advice on. 82 perinatal outcome of interventions. 18. 71 insulin initiation. 9 key research priorities. 15. 125 evidence description. 32 antenatal care. 107 16 weeks. see glucose (blood) monitoring individualised. 114 recommendations. 131 atorvastatin. 52 from evidence to recommendations. 109. 145 amniotic fluid erythropoietin level. 124 obesity effects. 125 angiotensin-converting enzyme (ACE) inhibitors. see glycaemic control targets. 74 gestational age for. 102 postprandial glucose levels (third trimester). 70. definition. 110 first appointment. 79 ultrasound measurement. 123. 109. see ACE inhibitors angiotensin-II receptor antagonists. xii antihyperglycaemic drugs. xii betamethasone. 123 recommendations. 110. 109. definition. 61. 110 recommendations. 125 blood glucose monitoring. 113 recommendations. 110 25 weeks. 51. 122. 66. fetal. 49 BNF recommendation. 145 audit. xii. 124 advice from healthcare providers. 110 18-20 weeks. 50 teratogenicity. 108. caesarean section. 112 current practice. 114 213 . 13. 50. 112 routine. 51. 100 American Diabetic Association. 20 timetable of appointments. xii alcohol intake. 109. poor pregnancy outcome frequency. glycosylated. 132 insulin. 20. 31 alpha fetoprotein (AFP). for women gestational diabetes. 144 congenital malformations associated. 106 anaesthesia. 16. 130 applicability. clinical. 113. 102 ultrasound measurement in gestational diabetes. 109. 63. 109. glibenclamide vs. 116 antenatal steroids evidence statement. 108 first appointment. see analgesia and anaesthesia (in labour) analgesia and anaesthesia (in labour). 15. recommendations. 54. 111 36 weeks. 39 albumin-to-creatinine ratio. 111 38 weeks. 95 in first trimester. 50 breastfeeding and. 53 in first trimester. 51 breastfeeding and. see also information. 114 insulin dose increase. 32 ACHOIS study. see also blood pressure for microalbuminuria. 74. 122. 113. 122. definition. 111 additional/specific for diabetes. 77 acarbose breastfeeding and. 15. see hypoglycaemic drugs. 52 teratogenicity. 52 recommendations. 124 from evidence to recommendations. 122 evidence statement. 52 from evidence to recommendations. 124 existing guidance. 109. see clinical audit Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS). 52 reference guide for use in pregnancy. 111 40 weeks. 110 28 weeks. gestational diabetes screening. 111 34 weeks. 122. 125 regional anaesthesia. 170 screening. 178 probabilities. reason. xii appraisal of evidence. 78 algorithm. 78. 110 antenatal risk factors. increased insulin needed. see also epidural anaesthesia acid–base status. 44 ACE inhibitors. 113 betamimetic drugs diabetic ketoacidosis and. 114 from evidence to recommendations. 173 amlodipine. 77 gestational diabetes treatment/screening cost-effectiveness. 84 for tocolysis. see also macrosomia large for gestational age (LGA). 71 treatment initiation. 95 albuminuria. 13. 94. 104 Apgar scores. 77 hypoglycaemia management. pre-existing diabetes. 144 gestational diabetes treatment. 15. 70. 124 research recommendations. 105. 50 antiobesity drugs. definition.

poor pregnancy outcome frequency and. 117 calcium-channel blockers breastfeeding and. 116 in pre-existing diabetes. 115 evidence statement. live vs stillbirth. 140 CEMACH enquiry. 117. 80. see drug safety. 146 initiation and maintenance. 97 causal relationship. 144 cardiac anomalies. 51 drugs whilst breastfeeding ACE inhibitors. 141 benefits. 137 barriers in diabetes. 138 obesity and IGT risk after. 146 insulin requirements. 144 statins. xii case series. 140 glycaemic control test. 51 calorie restriction. 60. 102 by third trimester non-fasting glucose levels. see also caesarean section in macrosomia. 122 risk associated with obesity and diabetes. 77 captopril. 134 birthweight and poor pregnancy outcomes. 40 hypoglycaemia during pregnancy. outcome. 143 recommendations. 13. 17. perinatal mortality. 150 pre-pregnancy counselling. 105 breastfeeding. 146 hypoglycaemia (maternal) prevention. 121 recommendations. 115 birth rates. 36 body weight. 29 folic acid use enquiry. 120 from evidence to recommendations. 138 supplementary milk/glucose. 117 from evidence to recommendations. 120 birth. 102 blinding (masking). intensive management effect. 118 birthweight. xii cardiotocograph. 121 optimal timing. 121 recommendations. xii case report (case study). 123 breastfeeding. 140.Diabetes in pregnancy bias. 122 routes and admission to neonatal intensive care. during breastfeeding early hypoglycaemia prevention. 34 poor pregnancy outcome. 15. 137. congenital. 143 obesity and IGT risk in infants. 78. 31 indications. 86 individualised antenatal care. 145 caesarean section antenatal risk factors. 121 existing guidance. see also congenital heart disease cardiomegaly. xii CEMACH. see also small for gestational age (SGA). see folic acid supplementation glucose testing in neonates. xvii performance. 61. see glycaemic control birth. 62 glibenclamide treatment. 81. xii gestational diabetes risk. 105 sub-optimal glycaemic control and poor outcomes. 141 glycaemic control (maternal). 97 quality of maternity service provision. 45 ACE inhibitor avoidance. see also macrosomia. 97. 51 obesity drugs. 145 obesity drugs. 81 congenital anomalies prevalence. 140 cow’s milk vs. 51 BNF recommendation for avoidance. 117 rates. 174 gestational diabetes vs pre-existing diabetes. xx birth injury. type 1 diabetes development. 115 increased. 2. 141 late neonatal neurodevelopmental impact. 10. 116 gestational glucose intolerance and. 146 oral hypoglycaemic agents and. 139 metformin excretion into milk. 17. 116 timing and outcome. 6 admission of neonates to intensive care. xix selection. 140 formula feeding vs. 121 gestational diabetes and. retinopathy reduction. 51 evidence statement. 130 routes and outcome association. 145 in pregnancy. 137. 137 frequent. 143. 143 evidence statement. 145 oral hypoglycaemic drugs. 79 by ultrasound. 128 sub-optimal glycaemic control in labour. glycaemic control during. macrosomia. 95 obstetric complications in pre-existing diabetes. 145. see also antihypertensive agents control. effect on. in macrosomia. prevalence.and micronutrients in breast milk. see glucose (blood) blood glucose targets. 120. 131 vs smaller babies. 53 retinal assessment in pregnancy. 119 diabetes effect on. 31 audit. 15. HbA1c in each trimester. gestational diabetes management. pre-existing diabetes. 143 evidence description. 115 vaginal birth after. 145. 72 increasing risk of diabetes after. 68. 116 epidural/spinal anaesthesia. 35 recommendations. see also caesarean section caesarean section vs expectant management. see weight (maternal) brachial palsy. 140 confidential demographic enquiry. 51 congenital anomalies associated. 30 postnatal care. 147 screening. hypoglycaemia during general anaesthesia. prevalence. xii caudal regression syndrome. 141 British National Formulary drug recommendations in pregnancy. 51 calcium-channel blocker avoidance. 141 drug safety during. 52 reference guide for use in pregnancy. 122 expectant management vs. 35 prediction. 138 macro. 91 body mass index (BMI). see glycaemic control targets blood pressure. 117 current practice. definition. 93 macular oedema and diabetic retinopathy. see vaginal birth vaginal birth outcome vs. acid–base status affected. 92 shoulder dystocia. 141 recommendations. 52 statin avoidance. 141 exclusive. 28 pre-pregnancy glucose tolerance status and. xii blood glucose. 91. gestational diabetes. 140. 138. xii information. 128 214 . see also hypertension. 145 calcium-channel blockers. see also large for gestational age (LGA) blood glucose levels relationship. 135 routes and intensive dietary management. 131 brain damage. timing/mode. 137 blood glucose levels. hypoglycaemia prevention. 143. 52 from evidence to recommendations. 107 nephropathy effect on pregnancy outcome. 137. 140. 119 planning of pregnancy and contraception use. 115 evidence description. 114 prevalence of congenital anomalies. 144 angiotensin-receptor blockers. 117 wound complication risk. see also vaginal birth. 6 congenital anomalies and stillbirth/neonatal deaths. 138 short duration vs early cow’s milk. 119. 133. 78 maternal weight affecting. cognitive/psychomotor development. xix publication. 115. 50 angiotensin-receptor blocker avoidance. 139. 141 ketone levels. 139. 56 removal of barriers to preconception care uptake. 145 from evidence to recommendations. 53 preterm labour in pre-existing diabetes. 115 in macrosomia. 96 descriptive study. factors affecting. 6 ethnicity and pregnancy outcome. xii case–control study. xvi. 116 routes and neonatal outcomes. 119 labour induction reducing. information for mothers.

37 preconception care decision tree. see congenital heart disease costs. 99 four-chamber view. 99 test characteristics. 165 test acceptance. 9. 184 215 . 179 cost–benefit study. 49 recommendations. 160. 2 confounder (confounding factor). 40 incidence. 187. 44 metformin. 41 first trimester HbA1c. 156 clinical trial. xiii Cochrane Library. xiv continuous subcutaneous insulin infusion (CSII). 55 use in type 1 vs type 2 diabetes. 28 safety of drug therapy for. see also health economics definition. 101 prevalence. poor pregnancy outcome. 169 sensitivity analysis. 95 obesity in pregnancy and. 194 timetable. see cost-effectiveness costs. 31 oral. 178 decision tree. 196 key recommendations. 159 prevalence by HbA1c levels by trimester. xiv in current guideline. xiii clinical questions. 101 ‘triple’ or ‘quadruple’ test. 9 model parameters. congenital. xix CNS anomalies. 163 professional time. diabetic. 15. xiii clinical importance. 179 risk factors within model. criteria. 171 screening. 195. 178. xiii complications of diabetes. 77. first-trimester screening. 191 decision tree. 107 centile. 51 insulin vs oral hypoglycaemic agents. 168 screening strategy assumptions. second-trimester scans. xiv congenital anomalies. 32 control group. 187 systematic review. 36. 96 benefits. diabetic pre-existing. 9 recommendations. 34 poor glycaemic control and. 165. 99 Cochrane Collaboration. xiv clinician. 98. see also randomised controlled trial (RCT) controlled. 99. 47 nephropathy in pregnancy and. 138. 43 risk to fetus in third trimester. 96 evidence statement. 30 CEMACH data. 192 model results. 194 outcomes and QALYs. 78. definition. 100 cost-effectiveness. 170. 184 prevalence as determinant. 30 evidence statement. 33 unplanned pregnancies and. 163 gestational diabetes screening. 195 current practice. 139 research recommendations. 100. 10. 98. xiv cost-effectiveness. 99 structural anomalies. 97 reduction by good glycaemic control first trimester. 167 model. 80 hyperglycaemia in early pregnancy. 45 in first trimester. pregnancy outcome. 74. 33 hypoglycaemia in pregnancy and. 36 cardiac. xiv consensus methods. xiv consistency. 47 insulin lispro and. xiii synthesis/appraisal of evidence for guideline. 110 transvaginal ultrasound. 18. 165 background to model. 98. 82 chlorpropamide adverse effects. 101 heart disease. 188 preconception care. 152 role/importance. 162. 97 congenital heart disease increased risk in diabetes. 32 from evidence to recommendations. see four-chamber view GDG recommendations. 32 ACE inhibitors. 54. 80 reduction by preconception care. 58 risk associated with drugs. 100 fetal echocardiography. 8 consensus statement. 191 biochemical marker levels. 99 types detectable. 21. 97. 141 cohort study. 80. 32 glycaemic control target. 139. xiv educational programmes on diabetes. xiii colostrum banking before birth. xiii Confidential Enquiry into Maternal and Child Health (CEMACH). 1. xiii. GDG members. 76. 186 costs. 18. 55 preconception care before discontinuation. xiii cluster randomisation. 166 modelling risk factors in screening. 33 key recommendations. 98 from evidence to recommendations. 6 three-part enquiry programme components. 44 oral hypoglycaemic drugs in first trimester. beliefs about. 168 screening strategy input parameters. 163 HbA1c levels in first-trimester. 81 CEMACH enquiry. poor pregnancy outcome frequency. 179 screening for congenital anomalies. 31 lack of use. see also neuropathy. 192. see CEMACH conflicts of interest. 6 timetable of antenatal appointments. 184 professional time costs for treatment. definition. 192. 44. 50 calcium-channel blockers. see congenital heart disease ICERs. xiii cephalopelvic disproportion. 172 outcomes. 141 benefits. xiii cognitive development. 191. 98 screening. see also retinopathy. 193 evidence description. 98 cilazapril. 166 cost analysis of treatment options. 101 research recommendations. 9. 5 clinical impact. definition. 51 screening. 22. see congenital anomalies consensus development conference. see also nephropathy. 100 existing guidance. xiii cohort. 9. 158 fetal growth and wellbeing monitoring. 21. 98 ultrasound. 101 second trimester. xiv cost(s) congenital malformations. 96. 162. 194 recommendations. 99 insulin glargine and. xiii. 144 clavicular fracture. breastfeeding vs cow’s milk. 142 macro. 57. see under insulin (therapeutic) contraception advice on. xiv. 45 secretion into breast milk. 131 clinical audit. 43 statins. 162. 101 congenital malformations. postprandial glucose monitoring. second-trimester. 144 chromosomal anomalies. 39. 150. xiii clinical effectiveness. 168 screening strategies. 105 gestational diabetes treatment/screening.and micronutrients. 139 comorbidity. 182. gestational diabetes treatment. 11 confidence interval (CI). 58 resumption after birth. 170 gestational diabetes treatment. 161. 107 survey. 195 self-monitoring of blood glucose.Index sub-optimal maternity care. 61. 100 first trimester. xiv controlled clinical trial (CCT).

40. 74 exercise with vs diet only. 32 Early Treatment Diabetic Retinopathy Study (ETDRS). 165 decision tree. 195 decision tree. 57. serum. 105 fetal aortic velocity waveforms. see fetal echocardiography 216 . 160 recommendations. 177 treatment results (from model). 29 pregnancy outcomes. 171 screening. 49 for diabetic complications. 162. 58. 176. 195. 87 risk with prolonged labour with epidural. see also individual types of drugs congenital anomaly risk. 187 preconception care. 86. see pregnancy outcomes prevalence. 185 treatment outcomes and downstream costs. 139 Diabetes Control and Complications Trial (DCCT). 68 gestational diabetes treatment. xiv decision analysis. 67 diabetes prevention after gestational diabetes. 103. diabetes risk after gestational diabetes. 68 gestational diabetes management. 176 treatment model parameters. xiv cost–utility analysis. 193 declaration of interest. 191 model parameters. definition. 145 reference guide for use in pregnancy. 191 discussion of results. see also folic acid supplementation preconception. 41 retinopathy progression and glycaemic control. 110 data set. 160.) treatment. xv dextrose. 67 insulin secretion/sensitivity. 58 double-blind study. 1. 104 Dose Adjustment for Normal Eating (DAFNE). 1. 14. 176 treatment decision tree. 33 evidence statement. 88 echocardiography fetal. 52 from evidence to recommendations. abdominal circumference. 11. 100. 86 drug safety. diabetic Diabetic Retinopathy Study. 10. 85 hyperemesis gravidarum. 58 research recommendations. in gestational diabetes. 158 costs. 35 diltiazem. 159 screening for congenital anomalies. 1 in pregnancy. 147 creatinine. 197 ICERs. 161 screening for congenital anomalies. 158 diabetes mellitus. see breastfeeding cranial-to-thoracic circumference ratio. 58. 67. xv Doppler ultrasound. 68 dietary supplements. 108 recommendation. 98 second-trimester ultrasound scan and. xv diet calorie restricted. see also type 2 diabetes. xv gestational diabetes treatment/screening glibenclamide treatment. 191. xv. 146 drugs. 149 pre-pregnancy. 74 duration before insulin. 161. 74 fasting blood glucose levels. active management of routes of birth and. 33. 13. 177 diet plus insulin vs diet only. 195 four-chamber view. 143. 104 umbilical artery cost-effectiveness. glucose intolerance association. 92 self-monitoring of blood glucose. 192. 76. xiv crown–rump measurements. before/during pregnancy. see nephropathy. 67. 159 discussion on model. 70. xiv cross-sectional study. 52 recommendations. 103. 49 evidence statement. 195 cost-effectiveness analysis. see also type 1 diabetes effect on pregnancy (maternal). 70. 19. 64 C-reactive protein (CRP). 31 Diabetes Prevention Program (DPP). see also under British National Formulary recommendations. effect on gestational diabetes risk. xv Down’s syndrome PAPP-A levels. 35 exercise with vs diet alone. 149 fasting. 11. 104 from evidence to recommendations. see also preconception care cow’s milk. 43 evidence statement. definition. 71 goal in pregnancies. 35 from evidence to recommendations. 106 fetal cardiotocography vs. 100. 48 recommendations. 101 screening. 36 recommendations. xv Delphi method. 67 intensive management. 167 treatment. 69 diet plus insulin vs diet only. 70 self-monitoring of glucose with. 163 decision tree. 116 low carbohydrate vs low fat/high in carbohydrates of low GI. 91 diagnostic study. see insulin and dextrose therapy neonatal hypoglycaemia. breastfeeding vs. 106 vasculopathy effect. 52 breastfeeding and. 9. 15. 29 management. 12. 28. 196 literature search and model. 31 preconception. 158 benefits. 21. 68 carbohydrate/fat/dietary fibre cross-over study. 161 sensitivity analysis of model. 33. 84 current best practice. 67 high GI diet vs. 178 treatment protocols. 40. xv screening/treatment of gestational diabetes. 67 poor pregnancy outcome frequency. 87 hospital stay and costs. 87 diabetic nephropathy. during breastfeeding. 89. 178 treatment options. 58 literature search. 98 Driver and Vehicle Licensing Agency (DVLA). 41 Diabetes Education and Self Management for Ongoing and Newly Diagnosed (DESMOND). 88 diabetic complications. 77 cost-effectiveness model. 165 counselling postnatal. 34 existing guidance. 191 costs. 31. 101. 17. 14. referral criteria in pregnancy. 180 neonatal care. 124 sequelae. 51 dominance (health economics). xiv screening for gestational diabetes. 59 crossover study design. 67 low glycaemic index.) gestational diabetes treatment/screening (cont. 135 preconception self-management programmes. 84 management. 36 weight-reducing. 53 drug safety. fetal wellbeing assessment. 33. 164 from evidence to recommendations. see complications of diabetes diabetic ketoacidosis (DKA). 104 velocity waveform analysis. aims. 177 risk factors. 69 evidence statement. 96 preconception.Diabetes in pregnancy cost-effectiveness (cont. intravenous insulin with. 58 results of model. 48 from evidence to recommendations. diabetic diabetic retinopathy. 194 sensitivity analysis. 159 quality-adjusted life years (QALYs). 28. 158 model parameters. see management of diabetes pregnancy affecting. diabetes prevention after gestational diabetes. 69 insulin therapy vs. 48 NSF for diabetes guidance. 33 advice. 192 model results. 11. 43 for diabetes management. see retinopathy.

105. 61. 120 fetal macrosomia. 35 experimental study. 132. 21. 105 current practice. audit of benefits. definition. 122. amniotic fluid. 102 amniotic fluid insulin levels and. xxi false positive. 53 type 1 vs type 2 diabetes. 69 effect on fasting glucose and insulin levels. 64. 136 economic evaluation. see also oral glucose tolerance test (OGTT) evidence description. fetal wellbeing assessment. 28 poor pregnancy outcome and. definition. 103 normal growth. xvi folate metabolism. 29 progression by pregnancy. 138 felodipine. 33 fetal size monitoring methods. see fetal growth and wellbeing first trimester. 28 poor pregnancy outcome. 76 advice to women. 60 diagnosis. 175 fasting plasma glucose (FPG) follow-up screening after gestational diabetes. xv evidence level. 103 fetal growth velocity and. poor pregnancy outcome frequency and. 60. 34 folic acid. 15. 149 gestational diabetes treatment. 102 ultrasound frequency. 105. 105 accuracy assessment. see also estimated fetal weight (EFW). 40 funnel plot. 106 frequency. 51 breastfeeding and. see information and follow-up after birth fosinopril. congenital heart anomalies. type 1 diabetes risk. 17. xv evidence-based clinical practice. 106 from evidence to recommendations. 100 ethnicity/ethnic factors gestational diabetes risk. preconception care. 34 evidence statement. 30 enalapril. xv epidural anaesthesia. 13. HbA1c) fluvastatin. xv for current guideline. 60 adverse pregnancy outcomes. unconjugated (uE3). 29. 35 preconception. 127 fetal echocardiography. 195 sensitivity and results. 144 four-chamber view. 53 pregnancy outcome and. 5 diagnostic tests. xvi fetal wellbeing. 124 glycaemic control and.g. 66 oral glucose tolerance test. 197 fractures. 132 estimated fetal weight (EFW). 29 European Surveillance of Congenital Anomalies (EUROCAT). optimal. 123. 135 at birth. xv empirical. 33. 166 feeding breastfeeding. hyperglycaemia in later pregnancy. xvi extrapolation. 69 glycaemic control improved by. 8 evidence based. xvi. see also Doppler ultrasound Doppler ultrasound. 34 follow-up after birth. 124 gastrointestinal anomalies. 145 fetal abdominal circumference. 36 from evidence to recommendations. 173 glycaemic control measurement. 102 cost-effectiveness. 95 energy supply. xvi. 106 research recommendations. 10. 36 regimes. 106 growth restriction. 115 clinical questions. 10. gestational diabetes screening. 102. 131 fructosamine test. 64 evidence statement. 103. hypoglycaemia risk. 104 fetal cranial-to-thoracic circumference ratio. 60. 35 dietary advice with vs alone. see breastfeeding early infant diet. 71 gestational diabetes. 105 growth velocity. 64 fetal distress. 28. 106 recommendations. xii best available. 35 recommendations. 193 outcomes. see macrosomia fetal risks. 156 definition. 99 fetal growth and wellbeing. during labour. definition. 123. 106 fetal heart monitoring. 144 end-stage renal disease. 102 evidence statement. 97 general anaesthesia. 57 education. Doppler ultrasound vs. see also abdominal circumference (fetal) accelerated growth. 65. 99 views used. gestational diabetes management. definition. 62 benefits of diagnosis/treatment. 66 GDG adoption of 75g OGTT. 131 erythropoietin. 73 educational programmes on diabetes cost-effectiveness. 145 focused question. 191 cost-effectiveness. 58 efficacy. 122 Erb palsy. 133. xv current guideline. 158 NICE recommendation. 105 existing guidance. congenital. 74 from evidence to recommendations. 77 appointment form for OGTT. 151 gestational diabetes screening. xvi false negative. 149. 66. in macrosomia. 51. 125 gestation. 33 epidemiology. maternal glycaemic control and. 36 use statistics. xvi folic acid supplementation absent. 62 recurrence. 103 macrosomia prediction. xv empowerment of women. xx family history (of diabetes). xvi gestational age admission to intensive/special care. see individual subjects (e. xvii interpretation by GDG. 69 diabetes prevention after gestational diabetes. 32. xv elective (procedure). 11. 99. definition. xvi experimental treatment. 76 diet with/without. xvi gastric emptying rate reduction. 105 type 1 vs type 2 diabetes. 33 wellbeing assessment methods. 107 ultrasound frequency. 105 evidence description. 36 NICE guidelines. 69 beliefs about in pregnancy. gestational diabetes vs controls. 35 evidence statement. 60 birth mode. 103. 62 screening. 124 acid–base status affected. 118 ultrasound assessment of abdominal circumference. definition. 197 costs. 64 217 . 35. 98. xv exercise barriers to in gestational diabetes. 68. see abdominal circumference fetal cardiotocography. definition. 94 accelerated progression in pregnancy. 195 decision tree. 36 recommended dose. definition. 117 body mass index relationship. xii hierarchy. 31 fetal surveillance. 115 birth timing. effects. 15. 103 recommendations. 7 economic model. xv. 19. 105 ultrasound reliability. 102 growth monitoring. 101 evidence appraisal. 7 evidence table. continuous electronic. xv.Index neonatal. 34 caesarean section and. 102 estriol.

75 in decision tree on screening. 69 calorie restriction. 165 economics model. 43 safety/low risk to fetus. 71. 13. 74. 61 systematic review. 165 type probability. cost-effectiveness. 151 likelihood/epidemiology. 18. 20. 137 supplementary for breastfeeding. modelling. 146 NSF for diabetes. 76. 76 cost-effectiveness model. 178. 165 risk factor-based. 77 systematic review. failed vaginal birth after caesarean section. see also insulin (therapeutic). 60. 177 education. 13. 76 maternity service providers using. 146 incidence comparison over time. 147 follow-up screening. referral criteria. 144 safety/low risk to fetus. 74 exercise. 146 costs. 75 methods to be avoided. 66 evidence statement. 179 QALYs. 177 insulin aspart vs human insulin. 1. cost-effectiveness. 151 vaginal birth after previous caesarean section. 184 self-monitoring of glucose. 12. 48 breastfeeding and. 120 screening. 62 recommendations. 144 in first trimester. 116 postnatal care.Diabetes in pregnancy gestational diabetes (cont. 73 in breast milk. 68 research recommendations. 144. 63. 46 intensive vs conventional. 118. 180 recommendations. 148 type 2 diabetes. 67 insulin analogues. 60. 75 indications for OGTT. see also diet. 144 safety/low risk to fetus. 13. 74. 59 glucose in breast milk. 187. 147 increasing risk. 44 glipizide breastfeeding and. 171 listing/summary. ICERs. 12 recurrence. 62 from evidence to recommendations. 173 BMI. 76 diet-controlled. 75 modelling of screening. 75. 22. 48 gliclazide. avoidance in pregnancy. 13. 185 oral hypoglycaemic agents. 151 lifestyle changes effect. 61 recommendations. 146. 61 follow-up. macrosomia risk. 76. see glycaemic control exercise effect in gestational diabetes. 168 Swedish study using OGTT. 177 outcome costs. 61. 170. 63 recommendations (criteria listed). outcome. 12. 62 evidence statement. 75 ICERs. 175 fasting blood glucose. 187 dietary. 73 effects on pregnancy outcomes. 128 immediate postpartum resolution. 170 age (maternal). 78 control/levels. 133 perinatal death association. 12. 117 glibenclamide as adjunct/alternative to insulin. 77 oral hypoglycaemic agents. 73 insulin analogues. 151. 12. 116 key research priorities. 60. 147. 44. 74. 62. 9. 78 metformin vs insulin. 72 cost-effectiveness. 148. 13. 67 glycaemic control protocol during labour. 177 decision tree. 60. 175 American Diabetic Association. 118 routes of birth and. 63 in cost-effectiveness for screening. 62. 77 intensive management effect on birth timing/outcome. 149. 44 glimepiride breastfeeding and. 66. 76 from evidence to recommendations. 18 macrosomia risk. 45. 77 rates and risk factors. 176. 177 insulin vs. safety. 43 insulin vs. 63. see also abdominal circumference (fetal) blood glucose targets and monitoring. 119 vaginal birth vs caesarean section. 165 test acceptance by women. 147. 170 BMI. 170. 63 family history. 66 types/modalities. 67. 172 previous gestational diabetes. 75. 62 ethnic groups. see cost-effectiveness ethnicity. 147. cost-effectiveness. 72. 44 gliquidone. 149. 184 global. 127 CSII vs glucose/glucose-free fluids. cost–utility analysis. 68. 20. 118 epidemiology. 61 cost-effectiveness. 176 costs. 66 accuracy of tests. 76 maternity service provider use of. calorie restriction effect. 18. 180 types/modalities. 47 insulin initiation. 76 neonatal admission to intensive care. 152 weight loss diet effect. 20 risk factors. 60. 44. see fasting plasma glucose (FPG) 218 . 77 information for informed decision.) recommendations. 70 type 1 or type 2 diabetes after. 186 cost-effectiveness. 12. 95 preconception period. 188 decision tree. 61 strategies. 12. 152 in obesity. 152 GI values of foods and.) diagnosis (cont. 12. see cost-effectiveness costs. 77 risk factor-based (only). 69 insulin lispro vs regular human insulin. 46 insulin glargine. 18. 148. 72 acarbose vs. 43 goal. cost-effectiveness. 76 research recommendations. 76 research recommendations. see also gestational diabetes. 76 cost-effectiveness model. 76 risk factor-based vs universal. see also gestational diabetes. 77 cost analysis of options. 77 dietary. 75 glibenclamide vs insulin. insulin analogue vs. 61 risk factor-based. 173 family history of diabetes. 61. 44 glomerular filtration rate (GFR) estimated. 172 sensitivity and cost-effectiveness. 18. 147 prevention by oral hypoglycaemics. 77 treatment. 140 glucose (blood) capillary (CBG). 165. 74 evidence description. 63 risks to infants. 61. screening age. 147 OGTT screening. 174 clinical characteristics and glucose tolerance. 69 fasting plasma (FPG). 62 recommendations. 72 BMI levels. 151. 20. 68. 149 prevention by lifestyle changes. 13. 150 prevalence. 71. 13. 189 gestational diabetes treatment. 68 information and follow-up after birth. 152 prevention by counselling and information. 76 pre-screening using risk factors. 75. cost-effectiveness. see above from evidence to recommendations. 69. 152 risk factors. 61. 75 ultrasound measurement of abdominal circumference. type 1 or type 2 diabetes after information needed by women. 44 options. 18. 13. 13. 17.

136 postprandial benefits. 11. 83 effectiveness. 9. 11. 93 improved control and retinopathy progression. 11. 83. 140. 79 preconception. see diet glycosylated albumin. 71 during pregnancy. 77 preconception. 113 birthweight relationship. pre-pregnancy. 81 existing guidance. breastfeeding effects. 10. 83 during pregnancy. 128 sub-optimal (in pregnancy) macrosomia. effects. 83 pre-eclampsia reduction. 79 macrosomia association. 70. 67 nausea and vomiting effect. see impaired glucose tolerance (IGT) glucose tolerance. 81 small for gestational age (SGA) and. 42 neonatal outcomes. 38. 33 planned vs unplanned pregnancies. 79 postpartum during breastfeeding. 80 recommendations. 129 research recommendations. low GI diet. 32 gestational diabetes diet vs diet plus insulin therapy. 18. see HbA1c (glycosylated haemoglobin) gold standard. see also hypoglycaemia (neonatal) breastfeeding vs formula feeding. 41 during anaesthesia. 179 diet with. 79 tight vs very tight control. third trimester. 125 maternal HbA1c in third trimester. 11. 139. 32 tight vs very tight. 36 first trimester. 31 postpartum. gestational diabetes management. 42 from evidence to recommendations. 79 preterm labour avoidance. 9. 140 maternal glycaemic control effect. 64 insulin lispro vs regular rapid-acting insulin effect. 69 in pregnancy. 56 planned vs unplanned pregnancies. 35 improvement. 124 exercise benefitting. gestational diabetes. 69 in pregnancy. 81 macrosomia prevention. 42 telephone modems. 83 research recommendation. 84 DCCT. 55 poor pregnancy outcome. 46 low GI diet effect. 81. see also HbA1c (glycosylated haemoglobin) fasting blood glucose. target in pregnancy. 78 breastfeeding effects. pre. 82 recommendations. 141. 108 intensive management in pregnancy. xvi glucose intolerance. 35 factors affecting. 84 pre. poor pregnancy outcome. see also oral glucose tolerance test (OGTT) body mass index relationship. 41 intensive treatment. 81 from evidence to recommendations. 82 respiratory distress prevention. 11. 83 third trimester. 81 optimising pre-pregnancy. 158 evidence statement. 11. miscarriage risk and. type 1 vs type 2 diabetes. see under breastfeeding preconception benefits. frequency. 41 existing guidance. 41 miscarriage risk. 127 testing. 18. 67 neonatal. 128 current practice. 32. 79 respiratory distress association. xvi growth. 33. 61 target ranges. 41 vs relaxed targets in labour. 14.Index fasting. 84 neonatal. see also HbA1c (glycosylated haemoglobin) congenital anomalies associated. 80 glycaemic control targets. 129 in gestational diabetes management. 82. see under breastfeeding during labour/birth. see also HbA1c (glycosylated haemoglobin) antenatal steroids affecting. 82 in gestational diabetes. miscarriage rate and. 54 low glycaemic index diet. 82 before going to bed. 42. 129 effect on neonatal glucose levels. 79. 41 glycaemic index (GI). 42 in gestational diabetes. 118 importance before pregnancy. 133 hypoglycaemia interventions. 28. 20. 79 recommendations. 127 from evidence to recommendations. 16. 82 from evidence to recommendations. recommendation. 41 evidence statement. 9. 14. 14. 129 infusion comparisons for. fetal. 39 glycosylated haemoglobin (HbA1c). 42 preconception. 92 effects of analgesia/anaesthesia on. 41 from evidence to recommendations. 89. 42 recommendations. 34 NSF for diabetes. 78 pre/postprandial glucose levels. 84 evidence statement. 83 recommendations. 82 pre-eclampsia. 120 knowledge about barrier to preconception care. 39.vs post-prandial. 67 glycaemic control. xvi. 135. 54. recommendations. 80 evidence statement. 128 insulin and dextrose therapy protocol.and postprandial levels. 40 testing. 60 hypoglycaemia definition.and post-prandial. 79 screening for gestational diabetes. 13. 80 individualised. 55 poor (preconception). 17. 14. birthweight predictor. macrosomia association. 81 macrosomia prevention. 128 existing guidance. 83 from evidence to recommendations. 55 poor pregnancy outcome. xvi grey literature. 83 targets. 79 glucose electrode. 14. 125 effect on retinopathy. 137 hypoglycaemia. 151. 38 intrapartum. 22. 42 self-monitoring after gestational diabetes in previous pregnancy. 42 preconception. 14. 40 ethnicity and. 30 in pregnancy. 69. 41 planned vs unplanned pregnancies. 129 recommendations. 38 recommendations. 18. 125 CSII vs glucose/glucose-free fluids. 125 frequency during labour. 32. 127 sub-optimal (in labour). 41 pre. 126 maternal choice of method. 82 lack. 83 existing guidance. CEMACH enquiry. 39. 142 non-fasting. 28 of DCCT. 129 neonatal hypoglycaemia. 14. 53 HbA1c. 122. 19. see hypoglycaemia (neonatal) optimal method?. see hypoglycaemia (neonatal) increased. recommendations. 78 congenital anomalies and stillbirth/neonatal death. 152 costs. see glycaemic control targets glucose (blood) monitoring continuous ambulatory. 81. 39 recommendations. 36. see fetal growth and wellbeing 219 . 20. 31. 81 poor pregnancy outcomes. 143 postprandial abdominal circumference and macrosomia. 89. see glycaemic control targets testing. 19. 36. 83 preconception. see also glycaemic control miscarriage risk reduction. 69 fetal outcome. 14. 84 research recommendations. 22. 128 fetal distress. 128 evidence statement. poor pregnancy outcomes. 82 gestational diabetes definition.

11. fetal. 2 declarations of interest. 125 effects. 78 preconception. decreased in pregnancy. see Guideline Development Group (GDG) development methodology. 31 macular oedema and diabetic retinopathy. 123. 131. 40 frequency. 41 levels for pregnancy avoidance. 177 in first trimester. 22. 133 hyperemesis gravidarum. 14. 137. 125 admission to intensive/special care. 6 recommendations formulation. 45 testing for. 136 breastfeeding. see also glycaemic control congenital anomalies association. 87 consequences. 86. 43 breastfeeding effect and safety. 143. xvi. expert advice. 120. glucose monitoring. 145. 139 persistent or recurrent. 139. 79 neonatal blood glucose levels and. 9. see breastfeeding current practice. 137. outcomes/risks. 17. 142 research recommendations. 74 hyperinsulinaemia. 8 health economic considerations. 142 intravenous dextrose for. 153 health economics. 86. 41. 140. 131 hypoglycaemia (maternal). see also other individual drugs. see also NSF for diabetes. neonatal. 91 third trimester macrosomia predictor. 82 prevalence by maternal glucose level. 134 postprandial glucose monitoring. 33 postprandial. 17. 62 problems associated with poor glycaemic control. 29. 142 evidence description. 8 research prioritisation. 40 congenital malformations and. 7 evidence levels for diagnostic tests. 123 HbA1c (glycosylated haemoglobin). 84 during pregnancy. 146 diabetes prevention after gestational diabetes. 17. definition. 60 hypertension. 40 gestational diabetes. 126. 48 aim and types. 57 interpretation of evidence. 40 pregnancy outcome and. recommendation. 43 220 . 7 priority. 136 definition. prioritisation of questions. 37 effect on miscarriage rates. 1 areas excluded from scope. see also entries beginning NICE recommendations. 3 updating schedule. 8 guideline(s). 28. 2 development group. 40. see recommendations haemodynamic control. 79 ultrasound markers for. 129. 136. xv. 6 stages. 146 risk with general anaesthesia. 82 health economics. 2 methodological support from NCC-WCH. 113 retinopathy progression and. 125 severe. 8 synthesis/appraisal of clinical effectiveness evidence. 140 intervention indication. 87 nocturnal. 14. 28 management. 42 neonatal outcomes. for diabetes in pregnancy. 40 awareness. 90 hypertension with. self-management programmes. 60 healthcare provider advice from. 14. xvii as outcome. 140 existing guidance. 47 management. 33 gestational diabetes. 132 characteristics. 73 increased in pregnancy. 140 early feeding effect. xx hypoglycaemia (neonatal). 40 pregnancy outcomes. 5 development. 33 Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study. audit of benefits. see statins homogeneity. 46 low GI diet effect. 8 membership. 4 stakeholder involvement. see also metformin advantages over insulin. poor pregnancy outcome frequency. 34 macrosomia and. 84 hyperglycaemia (maternal) continuous blood glucose monitoring during pregnancy. 91 retinopathy progression. 121 sulphonylureas causing. 86. 128 CSII vs intravenous insulin infusion effect. 158 self-management programmes. 83 early pregnancy. 54 heterogeneity. 56 planned vs unplanned pregnancies. 5. 46. see also blood pressure advice on. 71. 142 hypoglycaemic drugs. 28 prevention during breastfeeding. 90 hypocalcaemia. 11 gestational diabetes treatment. 55 preterm birth association. 140. 39 hypoglycaemia risk and outcomes. on diabetes in pregnancy) aim. 3 intended readership. gestational diabetes screening. effects. see also cost-effectiveness current guideline. 40 flexible intensive insulin treatment. 130 risk reduction by maternal glycaemic control. 8 specific considerations. 40 insulin lispro vs regular rapid-acting insulin. 87 frequency and risks. xvi. 6 literature search strategy. 126 intravenous glucose and insulin effect. 153 research prioritisation. 22. definition. 8 diagnostic test accuracy. 84. 140. xvii human chorionic gonadotropin (b-HCG). 177 decision tree. 127 not reliable indicator. 3 existing. 127 insulin and dextrose therapy protocol. 2 stakeholders. 38. recommendations. 69 insulin lispro vs regular human insulin. 141. 40 preconception care benefits. xvi clinical. 81. 4 maternal outcomes. 80 congenital anomalies association. 141. 39 recommendations. 140 frequent feeding effect. 74. 2 documents of relevance to. 8 Guideline Development Group (GDG). 6 neonatal outcomes. 2. 127 prevention/assessment. 125 risk factors. 42 use in CEMACH enquiry. 4 consensus methods. see self-management programmes (preconception) health technology. 133 CSII vs glucose/glucose-free fluids. 5 evidence levels. see advice from healthcare providers appointment form for OGTT. oral. 84 gestational diabetes treatment. 77 cost-effectiveness model. 7 GDG interpretation. 37 first trimester adverse pregnancy outcomes and. xvi health technology assessment (HTA). 152 discontinuation in pregnancy. 98 hyperbilirubinaemia. 11. 2 conflicts of interest. insulin vs glibenclamide. exercise with diet treatment. 29. 83 preconception. 7 intervention effectiveness assessment. 142 reasons for/pathogenesis. 9. 141 from evidence to recommendations. 141 recommendations. xvii. 17. 43. 66. 81 relationship with barrier to preconception care. 13. sequelae. 7.Diabetes in pregnancy guideline (this. 143. 136 evidence statement. 41 monthly. 76. 57. see also glycaemic control goal of DCCT and glucose monitoring. 14. 60 later pregnancy. xvii HMG CoA reductase inhibitors. xvi hierarchy of evidence.

48 insulin isophane (NPH insulin). xx intrapartum care. for women gestational diabetes management. 49 insulin detemir. 85 intravenous infusion vs during labour. after antenatal steroids. 73 intensive care. 18. 189 long-acting analogues. fetal.Index congenital anomalies risk. 19. 151 gestational diabetes. 47. 150 from evidence to recommendations. 9. 46 gestational diabetes treatment. gestational diabetes. 156 key research priorities. 133 risk in children born to mothers with diabetes. 48 insulin glargine. 85 effect on retinopathy. 34 treatment. abdominal circumference. 85 outcomes. 85 gestational diabetes treatment. 45. 10. 11. 11. 46 blood glucose levels. 177 glibenclamide vs. 47 proliferative retinopathy association. 73 metformin vs. 48 recommendations. 87 insulins to avoid in pregnancy. 49 oral hypoglycaemic agents with/without. 73 goals (pre-post-prandial glucose levels). 86 oral hypoglycaemic drugs advantage over. 71 insulin requirements and calorie restriction effect. 85 adverse effects. 47 regular human insulin vs. see analgesia and anaesthesia (in labour) clinical questions. 195. 86 recommendation. see gestational diabetes. 57 interquartile range (IQR). 196 individualised antenatal care. neonatal. 119 incremental cost-effectiveness ratios (ICERs) gestational diabetes screening. risk identification. 146 human. regular human insulin vs. 137. hypoglycaemia and. 86 effectiveness. see gestational diabetes preconception glycaemic control. preconception. 67 sensitivity. prevalence. 22. 143. 125 secretion/sensitivity. 44 hypoxia. 6 interview. 47 during pregnancy. 115 elevated amniotic fluid insulin levels. 129 protocol. 146 diabetes prevention after gestational diabetes. 70 diet vs. 150 recommendations. 45 recommendations. 87 multiple daily injection (MDI) regimen. 21. fetal distress. 77 glibenclamide costs vs. macrosomia risk. 107 221 . 113. 115 analgesia. timing/mode intrauterine death. adverse perinatal outcome reduction. 132 imidapril. 65 fetal cranial-to-thoracic circumference ratio. 185 preconception care. active management of routes of birth and. 118 lifestyle and drugs to prevent diabetes. see also insulin lispro. 45 gestational diabetes treatment. 152 information bias. 58 screening for congenital anomalies. see under neonatal care intensive management of diabetes conventional vs. 45. 74 cost-effectiveness model. 83 gestational. 85. 69 initiation based on fetal abdominal circumference. 143 with/without insulin. 48 safety assessment (in pregnancy). 145. 45 Dose Adjustment for Normal Eating (DAFNE). 45. 138 third trimester. 85. 48 congenital malformations and. 107 individualised glycaemic control targets. low GI foods and. see isophane insulin (NPH insulin) insulin lispro congenital malformations and. see labour induction information and follow-up after birth. 30 timing of provision. advantages. 74 in pregnancy. 69 diet only vs diet with. 48 supplementary infusion. 75. 85 insulin glargine vs. 48 safety in pregnancy. 43 NSF for diabetes guidance. 69 secretion in neonates. four-times daily vs twice-daily. insulin aspart vs. 114 continuous subcutaneous infusion (CSII. 45 before/during pregnancy. 49 placental transfer. 64 gestational. 57 inhibin. see also insulin aspart benefits in pregnancy. 48 recommendations. 19 research recommendations. 14. 86 benefits. recommendations. 47 pregnancy outcomes. 85 long-acting insulin vs. 47 regular rapid-acting insulin vs. low GI diets and. 146 evidence statement. 19. 43 insulin-meal intervals. glycaemic control. 47 insulin pump therapy. neonatal hypoglycaemia. 87 NICE guidance. 144 impaired glucose tolerance (IGT). 43 secretion into breast milk. 47 gestational diabetes treatment. 67. 73. 70 glibenclamide vs. regular insulin vs. 60 adverse pregnancy outcomes and. 48 human insulin vs. perinatal outcomes. xvii information. 122 insulin (therapeutic) breastfeeding and. 17. impaired by pain/stress. 127 new-generation. 46. 44 recommendations. 151 existing guidance. see birth. 149 evidence description. recommendations. 116 effect on long-term self-management and glycaemic control. 128 intravenous infusion vs during labour. see glycaemic control targets induction of labour. 44 rapid-acting analogues. 74 human. 20. semi-structured. 73. 11 recommendation in pregnancy. 73 in gestational diabetes cost-effectiveness. 13. 22 timing/mode of birth. xvii intervention definition. insulin lispro vs. 65 ultrasound fetal biometry and. 87 in pregnancy. 72. 86. 46. 13. 11. 85 increased by antenatal steroid use. 77 preconception. CEMACH. 71. 75 diet only vs diet with. 47 insulin glulisine. perinatal outcomes. see under insulin (therapeutic) insulin-meal intervals. 78 dietary. 40 gestational diabetes treatment. 68 risk of diabetes after. type 1 or type 2 diabetes after pre-existing type 1 or 2 diabetes. 47 systematic review. 47 insulins to avoid in pregnancy. 113 insulin analogues. 98 insulin (endogenous) amniotic fluid levels glucose intolerance and. see glycaemic control self-management and. 73. 66 caesarean section and. 79 initiation and indications for. 147 in pregnancy. 46. 46. 126 insulin aspart in gestational diabetes. 70. 147 postpartum reclassification to. 74 in pregnancy. 33 flexible treatment. fetal outcome. 64 evidence statement. 127 intravenous infusion vs during labour. recommendations. 73 in pregnancy. 103 exercise effect on levels in gestational diabetes. 73. 152 research recommendations. 66 in utero losses. 14. 11. see also insulin glargine in pregnancy. 45 insulin and dextrose therapy. xvii effectiveness assessment.

96 macrosomia. postpartum. CEMACH. 96 testing recommendations. 39 misoprostol. 117 evidence statement. 29 maternity service provision quality and availability. see also macrosomia abdominal circumference. 54 labour induction. xvii methodology. 48 gestational diabetes. 107 multiparous. 107 use of gestational diabetes screening. 94 thromboprophylaxis. 39. birth induction. 79 laser treatment diabetic macular oedema. 11. of diabetes control barrier to preconception care. xvii ‘medicalised’ labour/delivery. 117 capillary blood glucose levels and. definition. 78. 28. 79 third trimester. 146 methodological quality. 120 risk in pre-existing diabetes. 116 labour. 51. xiv ketoacidosis. xvii. 152 effectiveness. 73. xvii. 132 associated conditions. 70 HbA1c levels. 21. congenital. for current guideline. 6 Metformin in Gestational Diabetes (MIG) trial. 74 cost-effectiveness. 78 impaired glucose tolerance and. levels for. 151. 14. 28. 149. 75. in breast milk. 11. see glycaemic control lacidipine. 95 microaneurysms. glycaemic control during. 81 clinical vs ultrasound. 102 risk reduction. 82 from evidence to recommendations. 94. 120 judgement. 48 recommendations. 95. diabetic blood pressure effect. 87 intensive. 14. 91 laser treatment. oral during pregnancy. diabetic antihypertensive treatment effect. 48 intensive vs conventional management. 44 placental transfer. risks. 79 ultrasound markers. 121 existing guidance. 86 recommendations. effect on pregnancy outcome in microalbuminuria. 75. blood pressure control and. 42 from evidence to recommendations. 11. xvii multidisciplinary study. 74. 94. 117 MODY. 20. 115. 52. 144 literature review. xvii multidisciplinary team. 44 from evidence to recommendations. 103 poor pregnancy outcome and. 74. see intensive management of diabetes key recommendations. 144 multicentre study. xvii as adjunct/alternative to insulin recommendations. 79. 79 prediction estimated fetal weight. 18. 44 fetal/infant outcomes. 141 monitoring during pregnancy. xvii literature search. 64 insulin glargine and. 78 outcome of caesarean section. 44 metformin. 49 isradipine. 145 lifestyle factors diabetes development after gestational diabetes. 148. 78. 117 risk with/without diabetes. 56 planned vs unplanned pregnancies. recommendations. 64 increased plasma glucose linked. 83 recommendations. 92 management of diabetes. outcomes. 121 insulin-requiring diabetes. existing guidance. 78 definition. definition. 17. 95 pre-eclampsia prophylaxis. see also nephropathy. see breastfeeding large for gestational age (LGA). xvii methyldopa. 177 insulin vs. 42 testing strips. definition. 95 effect on pregnancy outcome. 105 postprandial glucose levels and. 30 maternal outcomes. 121 Mendelson syndrome. 6 adverse. xviii muscle relaxants. 102 admission to intensive/special care. 94 NICE definition. xvii ACE inhibitors in first trimester and. 11. 86 existing guidance. 73. 90. 95 microalbuminuria. 76 miscarriage. 116 poor glycaemic control and. 81 prevention. placebo vs. 149 poor pregnancy outcome. 31 lisinopril. in pregnancy. definition. 94. 123 musculoskeletal anomalies. see also hypoglycaemic drugs. abdominal circumference. 42 self-monitoring. 81 third trimester. 87 research recommendations. xix current guideline. 79 increased risk with expectant management. 102 risk in gestational diabetes. 120 MIG trial. 145 lactation. 90. 28. 75 maturity-onset diabetes of the young (MODY). as predictor for. 9. 102 estimated fetal weight vs abdominal circumference. 84 monitoring in preconception period. 123 meta-analysis. 41 evidence statement. 49 safety. see maturity-onset diabetes of the young (MODY) moexipril hydrochloride. 38. 144. 42 knowledge. 102 birth injury rate/risk. 78. see also abdominal circumference (fetal) admission to intensive/special care. 9 masking. see diabetic ketoacidosis (DKA) ketones breastfeeding vs formula feeding. 132 assessment methods. xvii median. support from. 131 lercanidipine. 4 longitudinal study. 31 regimens and impact. 41 preconception glycosylated albumin levels. 68 types (symmetric and asymmetric). 44. 76. retinal. 86 from evidence to recommendations. 78 macular oedema. NSF for Diabetes. 48 safety in first trimester. 102 detection. 120.Diabetes in pregnancy isophane insulin (NPH insulin) insulin analogue comparison. 133 222 . 56 sub-optimal (CEMACH enquiry). 131 caesarean section vs expectant management. 97 myocardial hypertrophy. see also insulin (therapeutic). 89 microvascular complications. 84 evidence statement. 121 macroalbuminuria. 143. 50 preconception glycaemic control effect. 120 calorie restriction in gestational diabetes. 43 metformin. 106 HbA1c levels associated. 92 proliferative diabetic retinopathy. 91 left ventricular outflow obstruction. 91 midwife. 113. considered. 52. 74. 19. 102. xvii lung immaturity. 116 optimal timing. 42 recommendations. see blinding (masking) maternal care. 116 effect on caesarean section rate. 137. 147. 120 from evidence to recommendations. 118 poor pregnancy outcome frequency and. 29 maternal risks. xvii biased results. 76 cost-effectiveness model. 84 evidence description. 145 jaundice.

120 neonatal intensive care indication. 95 self-management. 97 outcome. 135 current practice. 137. effects on analgesia/anaesthesia. see also renal assessment. 115 congenital anomalies not linked. adverse perinatal outcomes. recommendations. 130 from evidence to recommendations. 140 glucose monitoring during pregnancy. 178 increased risk in diabetes. 140 NICE technology appraisal. 48 postnatal care and follow-up. 134 conditions detected and signs of. 123 nominal group technique. 131 neonatal outcomes. 130 key recommendations. 80. 120 NICE Postnatal Care guideline. 165 follow-up and diabetes risk after. 75. 120 gestational diabetes and. 136 research recommendations. 35 recommendations. 135 neonatal morbidities. 157 cost-effectiveness. 134. 151 gestational diabetes. 130. 135 criteria/indications. 110 fetal growth and wellbeing monitoring. 60. 51. 35. 95 nephropathy testing. 138 obesity drugs breastfeeding and. 123 gestational diabetes. 65 appointment form. 35 glibenclamide vs insulin. definition. 66 recommendation. 81 in labour. 17. xviii adverse pregnancy outcomes. 75 indications. 106 first-trimester screening for congenital anomalies. 98 neurodevelopmental outcomes. xviii palsy. 124 microalbuminuria. 13. 130 criteria/indications. 124 nicardipine. NICE recommendations. 36 caesarean section risk. 98 second-trimester ultrasound screening. gestational diabetes treatment. 99. 31 effect on pregnancy outcome. 144 safety/low risk to fetus. 52. xviii observational study. 136 neonatal metabolic adaptation. audit on benefits. 48 maternal weight gain and. 36 risk in children born to mothers with diabetes. xviii. 138 neuropathy. 17. late neonatal ingestion of breast milk. see hypoglycaemic drugs. pregnancy outcome and. xviii NPH insulin. 95 existing guidance. adverse pregnancy outcomes. 107 early breastfeeding. 134 existing guidance. 77 abnormal. 86 nifedipine. 83 glycaemic control. 145 nitrous oxide. xviii glycaemic control target to prevent. 119 odds ratio (OR). 44 National Service Framework for diabetes. xviii nephropathy. 134 evidence statement. see isophane insulin (NPH insulin) NSF for diabetes. 148 risk factor-based indications. 6. 94. see also pregnancy outcomes adverse insulin glargine and. 51 nisoldipine. 84 negative predictive value (NPV). diabetic. 123. adverse. 136 recommendation to keep babies with mothers. 65 75g (2 hour) test. 134 from evidence to recommendations. 49 advice on. 11. 121 rate in diabetes. 130 admission to ICU or special care. 130. 130 perindopril. 9 recommendations. 116 incidence. 145 oral glucose tolerance test (OGTT) 100g (3 hour) test. 145 evidence statement. 128 intrapartum care. 95 pre-existing. 85 recommendation for educational programmes. 158 NICE Induction of Labour guidance. 62 follow-up screening after gestational diabetes. 132. 150 retinal assessment. 135 in type 1 diabetes (maternal). xviii nurse-led education. xix performance bias. 134 frequencies. 51 breastfeeding and. 95 testing. 39 insulin vs oral hypoglycaemic agents. 150 early breastfeeding. see hypoglycaemia (neonatal) initial assessment. 120 neonatal echocardiography. 120 NICE Intrapartum Care guideline. 81 rates. 134. xx neonatal care. poor pregnancy outcome frequency. 64 50g test. 52 breastfeeding and. 145 orofacial cleft. 120 NICE Diabetes guidelines analgesia and anaesthesia. 32 screening recommendation. 107. see NSF for Diabetes nausea and vomiting. 68 preconception evidence statement. 52 observation. see also microalbuminuria ACE inhibitor discontinuation in pregnancy. 52 in pregnancy. by HbA1c levels by trimester. xviii P value. 28 progression. 94 rate in pregnancy in type 1 vs type 2 diabetes. 66 antenatal care timetable. planned vs unplanned pregnancies. 3. 135 hypoglycaemia prevention. 130 incidence. 99 number needed to treat (NNT). 94. 136 clinical questions. 95 neural tube defect. 28. xviii olmesartan. oral orlistat. calorie restriction effect. 132. 136 neurodevelopmental outcomes. 16. 131 parenteral nutrition. 60 peer review. xviii non-experimental study. 51. 44 neonatal risks. 98. definition. xiii. xviii risk in diabetes. 22 neonatal death. 145 nimodipine. effect of pregnancy on. insulin pump therapy. 92 nuchal translucency test. neonatal hypoglycaemia. 41. 145 NICE continuous subcutaneous insulin infusion. in first trimester. 9. xix perinatal mortality. 149. health economics. xv. 64 cost-effectiveness. 123 nateglinide breastfeeding and. 10. 98. 64 as single test for gestational diabetes diagnosis. 100 NICE Caesarean Section guideline.Index narcotic analgesics. xv. see also vomiting effect on diabetes. Erb. 115. 73 obesity. 54 phenformin. 134 oral hypoglycaemic agent guidance. 84 Pedersen hypothesis. 131 evidence description. 62 oral hypoglycaemic drugs. 66. 28 neonatal unit. 43 glycaemic control effect. 34 effects on analgesia/anaesthesia. 96. 29 antenatal care timetable. 80 prevalence. 43 223 . 34. 34 hyperemesis gravidarum. 144 personality. 58 NICE Antenatal Care Guideline. xviii obstetric complications ACHOIS study. 130 signs and detection.

barrier to preconception care. 47 metformin. 81. 56 unplanned pregnancies. definition. 9 lack of formal arrangements for. importance. 30 pregnancy-associated plasma protein-A (PAPP-A). 57 key recommendations/priorities. 55 types and delivery forms. see pregnancy outcomes planned vs unplanned advice from healthcare provider. 10. 29 follow-up in pre-existing diabetes. 161. 53 planning. 17 research recommendations. 30 timing of provision. recommendations. 90 focal vs scatter. recommendations. see information and follow-up after birth key recommendations. 28. 28. 56 avoidance. ACE inhibitor discontinuation and. 31 improving. 10 type 1 diabetes vs no diabetes. 30 evidence statement. 21. 100 Down’s syndrome. 33 planning. 9 nephropathy effect on. 53 relationship with healthcare provider. methods. 54 prevalence. 120. 56 renal and retinal assessments. 158 clinical questions. 57 audit of uptake. 54 as barrier to preconception care. 163 decision tree. 66 retinal assessment unlikely. 56 planned. 21. xix 224 . xix placental insufficiency. 31 outcomes. chronic. 56 lack of social support. 120 recommendations. 76 HbA1c levels in first-trimester. 33 prevalence of live vs stillbirths. 19. 49 pregnancy. see renal assessment unplanned as barrier to preconception care. microalbuminuria testing and. 41 recommendations. 29 increased surveillance. 103 placental transfer insulin lispro. clinical questions. 55 knowledge about glycaemic control. 11. 45 placebo effect. 22 postnatal. see drug safety. see glycaemic control information for women. see diet drug safety. 31 relationship with partner. 54 without diabetes. 31 CEMACH enquiry into folic acid supplementation. see dietary supplements. 133 positive predictive value (PPV). 145 preconception care. see also diabetes mellitus avoidance. 57 barriers to availability/appropriateness of services. 156 components. 144 moderate risk to fetus. 58 benefits. weight and exercise. 83 gestational diabetes treatment types and. 54 diabetes vs no diabetes. 101 premature delivery. 56 lack of knowledge. 44. 160. poor pregnancy outcome frequency. definition. 143 breastfeeding and glycaemic control. 119 CEMACH enquiry into contraceptive advice. 94. see breastfeeding information and follow-up. 56 healthcare provider relationship. supplements. 158 most beneficial form and reasons. 56 characteristics/demography. 44. see pregnancy outcomes removal of barriers to uptake. xxi postnatal care. 41 beliefs about exercise. 30 advantages. 36 prediction by Doppler ultrasound. 19. 159 diet. see retinal assessment self-management. 54 personality and. 43 polycythaemia. 91 pilot study. see self-management programmes (preconception) summary of recommendations. 118 good. before/during pregnancy economic model. planned vs unplanned aims. 53 evidence statement. 29 effect on diabetes. 28. definition. 34. 103 prediction by Doppler ultrasound vs fetal cardiotocography. 58 cost-effectiveness. 56 glycaemic control. 31. 9. 53 availability. lack of. need for. see also pregnancy. 33 key recommendations. 34 CEMACH enquiry into glycaemic control tests. 131. 178 birth timing/mode. see pregnancy. definition. 58 young women. 57. 57 preconception clinic. see also cost-effectiveness cost savings with care/advice. see weight. 53 congenital malformation reduction. 9 recommendations. 28. 90 proliferative diabetic retinopathy. xix pioglitazone breastfeeding and. 92 risks related to diabetes. see retinal assessment. 91 no photocoagulation vs. 48 planning of pregnancy. 158. 119. 56 sociocultural factors. 108 pre-eclamptic toxaemia (PET). 82 pre-and post-prandial glucose monitoring. 58 costs. 157 pravastatin. HbA1c levels. 56. 31 CEMACH enquiry into social/lifestyle factors. 150 recommendations. 30 gestational diabetes and. 69 diabetes affecting. 32 CEMACH enquiry into pre-pregnancy planning. 11. 56. see preterm delivery/birth pre-pregnancy counselling. 32. 57 evidence statement. 53 planning of pregnancy and contraception. 54 relationship with partner. 57 retinal assessment. 54 poor outcome. 53 evidence. 33 contraceptive use. 150 from evidence to recommendations. 11. 55 pregnancy outcomes. see also impaired glucose tolerance (IGT) ACHOIS study. 79 prophylaxis. 162. 53 antenatal appointments after. 54 without diabetes. 56 from evidence to recommendations. 9 pre-existing diabetes. 51. 65. 98. see preconception care pre-eclampsia postprandial glucose monitoring. see antenatal care appropriateness of services. 57 before contraception discontinuation. see also congenital anomalies adverse. 10. see exercise. 31 preterm delivery/birth. 104 reduced by treatment of glucose intolerance. 57 recommendations. CEMACH. 28 sub-optimal glycaemic control in pregnancy. 55 evidence statement. 96 surveillance for. 57 glycaemic control. 32 from evidence to recommendations. 121 key recommendations/priorities. 10. 57. 29 evidence statement. 54.Diabetes in pregnancy photocoagulation diabetic macular oedema. 55 young women. 64 obesity and. 10. 80 levels of glucose intolerance. see contraception pregnancy outcomes. 95 of hyperglycaemia in early pregnancy. 31 ethnicity and. xix postpartum care. 56 young women. see under pregnancy polycystic ovary syndrome. xix placebo. 46. xix. 32 CEMACH data. 74 gestational glucose intolerance.

14. 88 insulin glargine and. 52 breastfeeding and. xx singleton. xix probability. xii. methodological quality-adjusted life years (QALYs). 88 postpartum regression. 93 methods. 51. definition. 86 duration of diabetes and. 113 rates. 28. 144 risk to fetus. 114 from evidence to recommendations. definition. 92. 59 preconception period. 94 evidence statement. 59 recommendations. labour induction and timing of. pregnancy outcomes and risks. 123 selection bias. xx self-management programmes (preconception). 95 during pregnancy. 96 testing for. xiii prospective study. 145 single-blind study. xix proteinuria ACE inhibitor discontinuation in pregnancy. 91 proliferative. xvii. 88 current practice. 132 assessment methods. 114 prevalence. 92 from evidence to recommendations. 105 postprandial glucose monitoring. 158. xx recommendations (guideline). 89 grading/classification. 59 retinopathy. 133 signs and detection. in pregnancy. 20. 82 prevalence rate. xv advice on. 90 hypertension effect. 93 continuous subcutaneous insulin infusion effect. 144 randomisation (random allocation). 92 HbA1c levels and. 93 through dilated pupils. see end-stage renal disease repaglinide avoidance whilst breastfeeding. xx sedatives. 91 duration of diabetes and. 113 evidence statement. xix question. 14. see cost-effectiveness research recommendations. 92 methods. 105 rate. 12. 121 risk reduction. preconception. xx rimonabant. 15. definition. xvii of study. clinical. 92 arrested by photocoagulation. 89 severity during pregnancy. 95 preterm labour. 89. 89. 96 research recommendations. poor pregnancy outcome. 31 blood pressure effect. 12. methyldopa use in microalbuminuria. 14. breastfeeding vs cow’s milk. xx shoulder dystocia. 113 current practice. 47 insulin lispro vs regular rapid-acting insulin. 80 recommendations. 9 preconception. management/follow-up. 130 myocardial hypertrophy and. 44. 30 relative risk (RR). 114 evidence description. xx rosiglitazone breastfeeding and. 93 retrospective studies. 113. 93 timing. 35 publication bias. see also glucose (blood) monitoring cost-effectiveness. 138. 88 evidence statement. xix gestational diabetes treatment. 89 type 1 vs type 2 diabetes. 120 admission to intensive/special care. 20. 119. 96 protocol. 95 preconception period. 91. 121 sibutramine. 93 research recommendations. 45 rosuvastatin. 79. 52 breastfeeding and. 8 summary. 130 reduction. 88. xx cohort study. xiii review. xix psychomotor development. definition. diabetes prevention after gestational diabetes. xx causal relationship. 89 factors associated. 14. 90 preproliferative. 19 respiratory distress syndrome. 133 incidence. definition. 88. 92 evidence description. recommendations. 14. xix quality methodological. 135 incidence. 96 renal disease. 89 glycaemic control effect. 160 screening for congenital anomalies. diabetic. xvi quinapril. 93 recommendations. 145 risk factors. end-stage. 12. 59 recommendations. xx sliding scale. 19 semi-structured interview. hypoglycaemia. xix ramipril. 8 key priorities. 180 preconception care benefits. 144 moderate risk to fetus. 21. 118. 14. 102 225 . focused. 89 insulin lispro and.Index admission to intensive/special care. 46 non-proliferative. xx sampling. 194 quantitative research. 132. 93 incidence and risk factors. 3. 144 quintile. 91 tight blood pressure control. xix prospective cohort study. 29. 93 severity at conception. 114 glycaemic control and. xx reliability. 79 risk. 29 risk ratio. definition. 15. 92 increased frequency of surveillance. 145 simvastatin. 49 pre-eclampsia surveillance. definition. 91 improved glycaemic control and. 44. xii range. xx renal assessment current practice. xx CEMACH enquiry. 108 referral criteria. 18 priorities set by GDG. 88 progression during pregnancy. 92 vaginal birth and. 88. 93 prevalence. see also individual topics formulation. xvi. xx Scottish Intercollegiate Guidelines Network (SIGN). xxi small for gestational age (SGA) admission to intensive/special care. 108 in gestational diabetes and pre-existing diabetes. 131 retinal assessment. xix randomised controlled trial (RCT). 132. 133. 90. 145 sample. 121 risk. xx during pregnancy. xvi. 91 referral. 149 public health guidance. 47 laser treatment. definition. 95 from evidence to recommendations. see quality. 141 psychosocial factors. 93. xx selection criteria. xiii question. 45 research recommendations key priorities. 78. 113 antenatal steroids administration. 59 recommendations. 92 existing guidance. 15. poor pregnancy outcome frequency. 88. xx sensitivity. proving. 106 postprandial glucose levels. definition.

6 World Health Organization (WHO). xxi after gestational diabetes. 45. 122 study checklist. 124 weight (maternal) adverse pregnancy outcomes and. of drugs in diabetes. antenatal. in macroalbuminuria. gestational diabetes. 105 information and follow-up (postnatal). xxi subject. 150 postnatal management plan. diet-controlled gestational diabetes. fetal wellbeing assessment. see congenital anomalies. 21. 126. 97 vaginal birth after previous caesarean section. 135 transposition of great arteries (TGA). for current guideline. 103 umbilical artery. 102 clinical estimates vs. 103 macrosomia markers. 120. 130 nephropathy and monitoring for. effect of diets on gestational diabetes risk. 197 trimester. see abdominal circumference birthweight prediction. definition. 29 maternal. 31 retinopathy. 35 influence on birthweight. 102 Doppler. 8 standard deviation. neonatal hypoglycaemia due to. 46 nephropathy and monitoring for. 80 prevalence. 119 caesarean section outcome vs. 52 from evidence to recommendations.Diabetes in pregnancy social deprivation ethnicity and. 83 teratogenicity. xxi systematic. 121 by HbA1c levels by trimester. 51 statistical power. 138 type 2 diabetes. 101 umbilical artery Doppler ultrasound. xxi startle reflex. 90. definition. 99. 191. congenital. folic acid supplements. lack of. 46 insulin requirements postpartum. 53 effect in pregnancy. 52 reference guide for use in pregnancy. see Doppler ultrasound fetal weight estimation. xxi thyroid function assessment. 17. 80. 56. 121 failed. short duration of breastfeeding vs early cow’s milk. 28 social factors. barrier to preconception care. 114 tolbutamide in first trimester. 33 reduction diets. definition. 150 preconception. 3 involvement in guideline development. 113. congenital anomaly detection. 144 trandolapril. systolic-to-diastolic ratio. gestational diabetes definition. 106 systematic error. 45 secretion into breast milk. 150 postnatal management plan. 32. 34 gain. 121 stress. 123 thromboprophylaxis. 131 statins. 135. 92 risk with early infant diet. HbA1c levels and. 3 NICE. definition. xxi systematic review. definition. xxi triple-blind study. 93 variables. 1. miscarriage risk and. adverse neonatal outcomes. 95 postnatal management. see fetal growth and wellbeing women. 118. 150 insulin lispro vs regular human insulin. 68 wellbeing. xxi spinal anaesthesia. 47 regular human insulin vs. xxi technology appraisal (TA). 58 telemedicine. definition. 28. 114 recommendations. 144 transfer of mother to obstetric unit. xxii causal relationship. recommendations. 91 ultrasound abdominal circumference measurement. 15. glucose monitoring during pregnancy. 95 postnatal management. 32 evidence statement. on structured education programmes. see methodological quality study type. 31 prevention by lifestyle changes. Doppler ultrasound of umbilical artery. 36 weight loss. 51 breastfeeding and. 60 wound complication risk. see also individual drugs. 120 transfer to community care. 57 risk-taking behaviour. 34 prevalence. 31 social support. 145 hypertension management in pregnancy. 145 congenital anomalies association. screening transvaginal. xxi symphysis–fundal height measurements. see antenatal steroids stillbirth. 78 screening for congenital anomalies. glycaemic control and. xxi sulphonylureas. 81 risk reduction. see gestational diabetes contraceptive use. 38. 117 young women preconception care and. folic acid supplements. 100 verapamil. 34 prevalence. 56 sociocultural factors. barrier to preconception care. 92 UK National Diabetic Retinopathy Laser Treatment. 150 preconception glycaemic control. 43 risk to fetus in third trimester. 35 preconception. 143 neonatal morbidities. in labour. see also nausea and vomiting regional anaesthesia associated. 44. 51 BNF recommendation for avoidance. see gestational diabetes congenital anomalies (second trimester). preterm labour. 11. xxi glycaemic control target to prevent. 118 gestational diabetes. 32 development. xxi. 117 diabetic retinopathy and. 114 from evidence to recommendations. of neonate. 45 survey. 56 226 . preconception. 95. 96 thrombosis. xii vasculopathy. xxi. 1. 99 contraceptive use. 136 transitional care unit. acid–base status affected. 150 insulin and dextrose infusion in labour. xxi study quality. 122 stakeholders. 138 fetal growth restriction. 51 breastfeeding and. 104 ventricular outflow tract scan. 46 insulin lispro. poor pregnancy outcome. 105 information and follow-up (postnatal). 14. 101. see also congenital anomalies thiopentone. 109 tocolytic agents. pregnancy outcome. 56 specificity. 197 detection rate threshold. 129 insulin aspart vs human insulin during pregnancy. 51 discontinuation in pregnancy. xxii after gestational diabetes. xxi study population. caesarean section. 104 urinary tract anomalies. xxi steroids. 51 vomiting. xxi target population. xxi type 1 diabetes. 151 NSF for diabetes. fetal. 151 NSF for diabetes. 149 retinopathy. 32 fetal growth restriction. 39 preconception.

.

To purchase further copies and for a complete list of RCOG Press titles.uk A version of this guideline for women with diabetes and the public is available from the NICE website (www.nice. Published by the Royal College of Obstetricians and Gynaecologists. treatment and long-term management Intrapartum care: care of healthy women and their babies during childbirth Atopic eczema in children: management of atopic eczema in children from birth up to the age of 12 years Surgical management of otitis media with effusion in children Guidelines in production include: • Induction of labour (update) • Surgical site infection • Diarrhoea and vomiting in children under 5 • When to suspect child maltreatment • Meningitis and meningococcal disease in children • Neonatal jaundice • Idiopathic constipation in children • Hypertension in pregnancy • Socially complex pregnancies • Autism in children and adolescents Enquiries regarding the above guidelines can be addressed to: National Collaborating Centre for Women’s and Children’s Health King’s Court Fourth Floor 2–16 Goodge Street London W1T 2QA enquiries@ncc-wch.com RCOG Press .uk/CG063) or from NICE publications on 0845 003 7783. quote reference number N1485.org.Other NICE guidelines produced by the National Collaborating Centre for Women’s and Children’s Health include: • • • • • • • • • • • • Antenatal care: routine care for the healthy pregnant woman Fertility: assessment and treatment for people with fertility problems Caesarean section Type 1 diabetes: diagnosis and management of type 1 diabetes in children and young people Long-acting reversible contraception: the effective and appropriate use of long-acting reversible contraception Urinary incontinence: the management of urinary incontinence in women Heavy menstrual bleeding Feverish illness in children: assessment and initial management in children younger than 5 years Urinary tract infection in children: diagnosis.org.rcogbookshop. visit: www.